CHARACTERIZATION OF A VANILLATE NON-OXIDATIVE DECARBOXYLATION GENE CLUSTER FROM STREPTOMYCES SP. D7

by

KEVIN TOSHIO CHOW

M.Sc, The University of British Columbia, 1996

B.Sc., The University of British Columbia, 1992

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

in

THE FACULTY OF GRADUATE STUDIES

(Department of Microbiology and Immunology)

We accept this thesis as conforming

to the required standard

THE UNIVERSITY OF BRITISH COLUMBIA

December 1999

© Kevin Toshio Chow, 1999 in presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.

Department of A//C&&I'QlOfr/ fflfJ* /MMtf^0Lcl 6rY

The University of British Columbia Vancouver, Canada

Date

DE-6 (2/88) ii

ABSTRACT

The genetics of non-oxidative decarboxylation of aromatic acids to phenolic compounds are poorly understood in both prokaryotes and eukaryotes. Although such reactions have been observed in numerous microorganisms acting on. a variety of substrates, genetic analyses of these processes have not, to my knowledge, been reported in the literature.

Previously, I isolated a streptomycete from soil (Streptomyces sp. D7), which efficiently converts 4-hydroxy-3-methoxybenzoic (vanillic) acid to 2-methoxyphenol (guaiacol).

Protein two-dimensional gel electrophoresis revealed that several proteins are synthesized in response to vanillic acid, one of which was characterized by partial amino-terminal sequencing, leading to the cloning of a gene cluster from a genomic lambda phage library of Streptomyces sp. D7. This cluster consists of four open reading frames, vdcA

(sequencing in progress), vdcB (602 bp), vdcC (1424 bp) and vdcD (239 bp). Protein sequence comparisons suggest that the product of vdcB (201 aa) is similar to phenylacrylate decarboxylase of yeast; the putative products of vdcC (475 aa) and vdcD

(80 aa) are similar to hypothetical proteins of unknown function from various microorganisms, and are found in a similar gene cluster in Bacillus subtilis. VdcA is a putative transcriptional regulatory gene. VdcB, vdcC and vdcD homologues are also clustered, along with putative />-cresol methylhydroxylase and vanillin oxidoreductase genes, on the 184 kb catabolic plasmid pNLl of Sphingomonas aromaticivorans F199.

Northern blot analysis revealed the synthesis of a 2.5 kb mRNA transcript, which hybridized strongly to a vdcC gene probe, in vanillic acid-induced cells, suggesting that the cluster is under the control of a single inducible promoter. Expression of the entire vdc gene cluster in Streptomyces lividans 1326, as a heterologous host, resulted in that Ill strain acquiring the ability to decarboxylate vanillic acid to guaiacol non-oxidatively.

Both Streptomyces strain D7 and recombinant S. lividans 1326 expressing the vdc gene cluster do not, however, decarboxylate structurally similar aromatic acids, suggesting that the system is specific for vanillic acid. By Southern blot hybridization, we detected the presence of the vdc gene cluster in several streptomycetes, including Streptomyces setonii

75Vi2, which has been previously shown to decarboxylate vanillic acid in a non- oxidative reaction. The vanillate decarboxylase catabolic system may be useful as a component for pathway engineering research focused towards the production of valuable chemicals from forestry and agricultural byproducts. TABLE OF CONTENTS

ABSTRACT ii

TABLE OF CONTENTS iv

LIST OF TABLES vii

LIST OF FIGURES viii

LIST OF ABBREVIATIONS xi

ACKNOWLEDGEMENTS xii

Chapter 1: INTRODUCTION 1

1.1 Non-oxidative decarboxylation: an industrially useful but poorly

characterized process 4

1.2 Non-oxidative decarboxylation in prokaryotes 7

1.3 Structure and function studies of non-oxidative decarboxylases 9 1.4 Metabolic engineering for the production of chemicals from renewable resources 10

1.5 Streptomyces: versatile soil microbes 11

1.6 Isolation and identification of Streptomyces sp. D7 15

1.7 Catabolic tests and 2D-PAGE analysis 17

1.8 Objectives 22

Chapter 2: MATERIALS AND METHODS 23

2.1 Bacterial strains and plasmids 23

2.1.1 Bacterial strains 23

2.1.2 Isolation and cloning of plasmids 23

2.1.3 16s rDNA sequence-based strain identification 28

2.2 Media and growth conditions 28

2.3 Library construction and gene cloning 31

2.3.1 Lambda phage DNA preparation 31 2.4 DNA sequencing and analysis 32

2.5 Southern blotting and hybridization 33

2.6 Chemical analyses 34

2.7 RNA isolation and analysis 34

2.8 Gene expression 38

2.8.1 pET22b(+) - E. coli BL21 38

2.8.2 pIJ702 - S. lividans 1326 38

2.8.3 pIJ680 - S. lividans 1326 40

2.9 Enzyme assays 42

2.10 Recombinant protein purification from E. coli BL21(DE3) 42

2.10.1 Ly sate preparation 43

2.10.2 Inclusion body protein purification,

solubilization and refolding procedure 43

2.10.3 Soluble protein purification 44

2.11 Construction of a knockout allele of the vdcC gene 44

2.12 Streptomyces sp. D7 protoplast preparation and transformation 49

2.13 E. coli-Streptomyces interspecies conjugation 49

2.13.1 E. coli S17-1 /pPM801 preparation 49

2.13.2 Streptomyces sp. D7 preparation and mating 50

2.14 Reagents and enzymes 5 0

Chapter 3: RESULTS 51

3.1 16s rDNA sequence identification 51

3.1 Cloning of the VDC gene cluster 51

3.2 Detection of a putative transcriptional regulatory gene 69 3.3 Messenger RNA analyses 72

3.4 Detection of vdc genes in S. setonii 75Vi2 74

3.5 Gene knockout studies 76

3.6 Gene expression 77

3.6.1 pET22b(+) - E. coli BL21 77

3.6.2 pIJ702 - S. lividans 1326 81

3.6.3 pIJ680 - S. lividans 1326 81

3.7 Substrate specificity 82

Chapter 4: CONCLUSIONS 88

Chapter 5: DISCUSSION 89

5.1 Aromatic acid non-oxidative decarboxylases are

multi-subunit enzymes 89

5.2 Distribution of the VDC gene cluster among streptomycetes 91

5.3 Substrate specificity 92

5.4 Primary structure motifs 93

5.5 Transcriptional activation 93

5.6 Biodegradation - a result of the microbial community gene pool 94

5.7 Sphingomonas aromaticivorans F199 catabolic plasmid pNLl 96

5.8 Metabolic engineering applications for vanillate decarboxylase 103

5.9 Future directions 106

Chapter 6: LITERATURE CITED 109 vii

LIST OF TABLES

Table Page

1 Bacterial strains and plasmids 24

2 PCR primers and hybridization oligonucleotides 36

3 Variations in subunit size and configuration - characteristics of 90

some microbial aromatic acid non-oxidative decarboxylases viii

LIST OF FIGURES

Figure Page

1 Lignin backbone structure 2

2 Solubilization of lignin yields a mixture of low-molecular weight 3 aromatic compounds

3 Reaction scheme for non-oxidative decarboxylation of vanillate to 6 guaiacol

4 The Streptomyces growth cycle 14

5 A possible niche for Streptomyces in lignin degradation 15

6 Decarboxylation of vanillic acid to guaiacol by Streptomyces sp. D7 19

7 Experimental scheme for growth and vanillic acid induction of 20 Streptomyces sp. D7 cells for 2D-PAGE analyses

8 Synthesis of protein 3717 in response to vanillic acid 21

9 Map of expression vector pIJ680 29

10 Map of cloning/expression vector pi J702 30

11 pIJ702 and the 4 kb BamHl Streptomyces sp. D7 genomic DNA 39 fragment inserted in both orientations in the tyrosinase gene (mei) promoter

12 pIJ680 and the PCR-generated gene combinations inserted 41 downstream of the promoter for the aminoglycoside phosphotransferase (aph) gene

13 Strategy for a knockout of the vdcC gene using the thiostrepton (tsr) 45 resistance gene

14 Amino terminal sequence alignment of Streptomyces sp. D7 protein 54 3717 and Clostridium hydroxybenzoicum /(-hydroxybenzoate carboxy-lyase

15 Lambda DASH II phage clone carrying the putative vanillate 55 decarboxylase gene

16 DNA purified from phage clone 3717C(+) and digested with BamHl 56

17 pSUBl (Lane A), pSUB3 (Lane B), pSUB4 (Lane C), pSUB5 (Lane 57 D) and pSUB6 (Lane E) purified plasmid DNA, digested with BamHl 18 Plasmids carrying subfragments of the BamHI genomic DNA 58 fragment from Streptomyces sp. D7, which was cloned in lambda DASH II phage clone 3717C(+)

19 Nucleotide sequence of the vdc gene cluster, featuring vdcB, vdcC 59 and vdcD

20 Schematic diagram of the 4.4 kb BamHI genomic DNA fragment 60 from Streptomyces sp. D7, containing the VDC gene cluster

21 Dendrograms depicting the sequence-based relationships between 61 proteins and hypothetical proteins similar to the product of the vdcB gene and vdcC gene

22 Location of the putative divergent regulatory gene in relation to the 70 VDC gene cluster

23 BLAST-X result indicating that the nucleotide region immediately 71 upstream of vdcB encodes a polypeptide similar to a putative transcriptional regulator from S. coelicolor A3 (2)

24 Northern blot hybridization of PCR-amplified, radiolabeled vdcC 73 DNA against mRNA isolated from uninduced Streptomyces sp. D7 (Lane 1) and Streptomyces sp. D7 induced with 3.6 mM vanillic acid (Lane 2)

25 Southern blot hybridization of Sa/I-digested chromosomal DNA 75 from Streptomyces sp. D7 (Lane 1) and Streptomyces setonii 75Vi2 (Lane 2) with radiolabeled vdcB, vdcC and vdcD PCR-amplified DNA probes at 65 °C

26 SDS-PAGE (12.5% acrylamide) of cell extracts showing the 79 expression product of the vdcC gene, produced using pET22b(+) in E. co//BL21(DE3)

27 SDS-PAGE (12.5% acrylamide) of cell extracts showing the 80 expression product of the vdcC gene, produced using pET22b(+) in E. coli BL21 (DE3) at 25°C and 30°C with 0.1 mM IPTG for induction

28 Decarboxylation of vanillic acid to guaiacol by recombinant 84 Streptomyces lividans 1326 strains

29 Results of incubating the insoluble and soluble fractions of 85 sonicated cell extracts of S. lividans 1326 - pKCS8 with 1 mM vanillic acid for fifteen minutes at 25°C under both aerobic (O2) and anaerobic (N2) conditions X

30 Expression of the vdc genes under control of the aph promoter in 86 pIJ680

31 Chemical structures of vanillic acid and similar aromatic acids used 87 to test the substrate specificity of vanillate decarboxylase

32 Amino acid sequence comparisons between the Streptomyces sp. D7 97 vdcB translation product (strPAD) and its Sphingomonas pNLl homologue (sphPAD) (a), and the Streptomyces sp. D7 vdcC translation product (strC) and its Sphingomonas pNLl homologue (sphC)(b)

33 Physical map of pNLl and the location of the VDC gene cluster 102 homologues

34 Vanillate decarboxylase, Cytochrome P-450 and catechol 1,2- 105 dioxygenase can be used in combinations to produce various industrially useful chemicals from vanillic acid as a starting material

35 Non-oxidative decarboxylases represent connections between the 107 two major branches of aromatic acid catabolism, characterized by either catechol or protocatechuate central intermediates LIST OF ABBREVIATIONS

Abbreviation Definition 2D-PAGE Protein two-dimensional polyacrylamide gel electrophoresis aph Aminoglycoside phosphotransferase gene BLAST Basic Alignment and Search Tool DNA Deoxyribonucleic acid DTT Dithiothreitol EDTA Ethylenediaminetetraacetate FAD Flavin adenine dinucleotide FAME Fatty acid methyl ester HPLC High pressure liquid chromatography Kb Kilobase KDa Kilodalton mei Melanin synthesis gene (tyrosinase) MRNA Messenger ribonucleic acid MSMYE Mineral salts medium with yeast extract PAD Phenyl acrylate decarboxylase PCR Polymerase chain reaction PEG Polyethylene glycol pi Isoelectric point NMR Nuclear magnetic resonance RBS Ribosome binding (Shine-Delgarno) site RNA Ribonucleic acid SSC Sodium saline citrate SDS Sodium dodecyl sulfate TSB Trypic soy broth UV Ultraviolet VDC Vanillate decarboxylase YEME Yeast extract malt extract Xll

ACKNOWLEDGEMENTS

I thank all the members, past and present, of the Julian Davies Laboratory for their support, encouragement, and friendship. In particular, I thank Rumi Asano, Dr. Jeff

Rogers, Sakura Iwagami-Hayward, Grace Law and Richard Kao for being great graduate school companions. In addition, I gratefully acknowledge Dr. Margaret K. Pope for providing essential guidance and technical expertise for the project. I thank Dr. Vera

Webb for making it possible to prove myself in the laboratory. Vincent Martin, Professor

William Mohn, Professor Douglas Kilburn, and Professor James Kronstad provided valuable technical advice for the project. I am grateful to my parents, who always believed in me, and Alisa Chan, who provided much appreciated support to get me through the most trying stages of graduate school life. Canadian Forest Products

(Canfor) Research and Development Centre staff provided valuable expertise and resources. Most importantly, Professor Julian Davies (who said I would be rich and/or famous someday) provided me with the encouragement, support, finances, and worldwide network of contacts that made my efforts worthwhile. Finally, there are many others who helped me in one way or another, and to them I am sincerely thankful.

This research was funded by the GREAT Scholarship program provided by the Science

Council of British Columbia (SCBC), Forest Renewal British Columbia (FRBC), and the

Natural Sciences and Engineering Research Council of Canada (NSERC). 1

1. INTRODUCTION

Chemical manufacture of benzenoid compounds from petroleum relies on abiotic, chemical catalysts. The use of petroleum poses a number of problems, it being a non• renewable resource, a geopolitically volatile commodity, and a source of many environmentally toxic compounds. Therefore, there is growing interest in developing processes for enzymatic conversion of renewable resources such as plant biomass for the production of chemicals traditionally derived from petroleum. The abundance of phenylmethylether motifs in natural compounds such as lignin (Figure 1) and their release as a result of lignin solubilization (Figure 2) has resulted in the evolution of mechanisms for the degradation of phenolic structures by microorganisms.

Microorganisms exhibiting such enzymatic biotransformation potential could be harnessed for industrial use to supplement or replace traditional chemical synthesis methods, should the need arise (Frost & Draths, 1995).

Vanillic acid is an abundant component of solubilized lignin biomass, and degradative mechanisms by which this compound is catabolized have been elucidated in several prokaryotic organisms. The genes responsible for vanillate demethylation have been cloned and sequenced from Pseudomonas sp. strain ATCC 19151 (Brunei & Davison,

1988), Acinetobacter sp. ADP1 (Segura & Ornston, 1997) and, most recently,

Sphingomonas paucimobilis (Nishikawa et al, 1998). In these microbes, vanillate is converted to protocatechuate, which is in turn degraded by enzymes of the P-ketoadipate pathway. CH2OH I

HC CH2OH "°"

Figure 1: The lignin backbone structure. 3

CH2OH

Figure 2: Solubilization of lignin yields a mixture of low-molecular weight aromatic compounds, some of which are shown here. 4

However, in some strains of Streptomyces and Bacillus (Crawford & Olson, 1978), vanillate is catabolized via an alternative pathway involving non-oxidative decarboxylation to guaiacol (Figure 3), with further catabolism via cytochrome P-450 mediated demethylation and mineralization through the intermediate catechol. In fact, it was demonstrated that individual Streptomyces isolates degraded vanillate by both routes, that is, through both catechol and protocatechuate as central intermediates. There are a number of reports in the literature of aromatic acid non-oxidative decarboxylases from various microorganisms (Grant & Patel, 1969; Yoshida & Yamada, 1985; Nakajima et al, 1992; Huang et al, 1993; Santha et al, 1995; He & Wiegel, 1995; He & Wiegel,

1996; Zeida et al, 1998) but thus far there have been no molecular studies of these processes.

1.1 Non-oxidative decarboxylation: an industrially useful but poorly

characterized process

Non-oxidative decarboxylation of aromatic acids involves the removal of the carboxyl moiety from the benzene nucleus via a reaction that requires neither oxygen, nor

cofactors such as NAD and FAD, typical elements of the oxidative process. The non-

oxidative process results in a "clean" removal of the carboxyl group, in contrast to the

oxidative reaction, which substitutes a hydroxyl group at the relevant carbon atom. In

nature, non-oxidative decarboxylation is not only observed for biodegradative pathways,

but also for anabolic pathways, such as in the biosynthesis of naphthoquinones (Santha et

al, 1995). Similarly, for biotransformation and metabolic engineering applications, both

oxidative and non-oxidative processes are valuable as components of hybrid pathways

for the production of various industrially useful compounds. In fact, a Klebsiella 5 aerogenes protocatechuate non-oxidative decarboxylase has been engineered into a hybrid pathway to produce catechol, a useful building block for pharmaceuticals, from glucose as a renewable starting material (Frost & Draths, 1995). 6

COOH

Vanillic Acid Guaiacol

Figure 3: Reaction scheme for the non-oxidative decarboxylation of vanillate to guaiacol. 7

1.2 Non-oxidative decarboxylation in prokaryotes

Reports on microbial non-oxidative decarboxylation of aromatic acids date back as far as

1924, but the first published work that was of significance towards modern studies of these enzymes was in 1969, by Grant and Patel, then at the University of Nottingham.

Their thorough physiological study of Klebsiella aerogenes and its abilities to non- oxidatively decarboxylate />-hydroxybenzoate, gentisate, protocatechuate and gallate revealed an organism that was likely responsible for decarboxylations of phenolic acids found in intestinal microflora. The study suggested that the aforementioned aromatic acids were decarboxylated by different enzymes, and that all of these enzymes were membrane associated, according to activity localization after ultrasonication and debris fractionation. The researchers noted that the non-oxidative decarboxylation of p- hydroxybenzoate was likely an injurious side reaction to the bacterium due to the toxic characteristics of the product, phenol.

As already mentioned, and relevant to this study, Streptomyces is one of the microorganisms in which non-oxidative decarboxylation has been biochemically well characterized. Crawford and Olson published data that would become the first in a series of reports detailing a streptomycete's capability to catabolize vanillate by decarboxylation to guaiacol (Crawford & Olson, 1978). Streptomyces strain 179 was isolated from Idaho forest soils, and exhibited what was then considered a novel catabolic reaction for the utilization of vanillic acid. Further research on another microorganism,

S. setonii 75Vi2, revealed that, in addition to non-oxidative decarboxylation of vanillate to guaiacol, a cytochrome P-450 system is involved in demethylating guaiacol to 8 catechol. Catechol is then mineralized (presumably, though not demonstrated) by catechol 1,2-dioxygenase and other associated lower pathway enzymes (Sutherland,

1986). Although non-oxidative decarboxylation of vanillate by S. setonii 75Vi2 was well-documented biochemically, the genetic basis for the reaction was not investigated.

The purification and partial sequencing of two non-oxidative decarboxylase enzymes from the strict anaerobe Clostridium hydroxybenzoicum has been accomplished (He and

Weigel, 1995). C. hydroxybenzoicum, an novel anaerobe, was isolated from freshwater sediments. The microorganism decarboxylates both 4-hydroxybenzoic acid and 3,4- dihydroxybenzoic acid under anaerobic conditions but does not metabolize the products further. The enzymes responsible for these reactions, />-hydroxybenzoate carboxy-lyase and 3,4-dihydroxybenzoate carboxy-lyase, showed no significant similarity to any protein sequences in the databases. The authors indicated that cloning of the genes encoding these proteins was in progress, but at the time of writing of this thesis, no such gene sequences have been reported.

From an industrial standpoint, non-oxidative decarboxylases that convert gallate to pyrogallol are particularly interesting. There are numerous reports in the literature detailing whole cell bioconversions of gallate using various microorganisms, both prokaryotic and eukaryotic. Yoshida and Yamada (1982) described the optimization of whole cell bioconversions using a Citrobacter sp., in which a yield of 97.4% pyrogallol

was obtained from gallate as an initial substrate. A recent development with great

industrial potential was a report of the purification and characterization of gallate 9 decarboxylase from Pantoea agglomeransTll (Zeida et al, 1998). P. agglomeransTll expresses not only a gallate non-oxidative decarboxylase, but also a tannase, which allows the organism to produce pyrogallol from tannic acid, an abundant waste product of the forest industry. The gallate decarboxylase was highly specific for gallate. As in the other studies mentioned in this survey, the genes encoding the decarboxylase were not cloned; however, the study emphasized the industrial relevance of this seemingly simple, yet poorly characterized, reaction.

1.3 Structure and function studies of non-oxidative decarboxylases

Several reports have described the instability of aromatic acid non-oxidative decarboxylases during purification procedures. Perhaps it is this characteristic of these enzyme systems that has resulted in limited studies of the proteins and their corresponding genes. Nevertheless, several research groups have succeeded in purifying, partially sequencing, and characterizing various non-oxidative decarboxylases. Santha et al. (1995) reported the structure and function of 2,3-dihydroxybenzoic acid decarboxylase from Aspergillus niger. In this work, the active site peptide of the enzyme was determined, and a partial primary structure map was created based upon sequences derived from enzymic cleavage of the protein. The enzyme system did not require any cofactors, emphasizing the importance of multiple active site residues (yet to be characterized) in the reaction mechanism. Huang et al, in a report describing the mechanism of action of ferulate decarboxylase in the yeast, Rhodotorula rubra (the enzyme does not decarboxylate the aromatic nucleus, but rather a side chain of the compound), also observed vanillate decarboxylation to guaiacol as a downstream 10 catabolic step (Huang et al, 1993). The researchers included a NMR analysis of the chemical mechanism of vanillate non-oxidative decarboxylation in their studies. He et al, in their study of the non-oxidative decarboxylases from C. hydroxybenzoicum, found that the enzyme did not require any cofactors or metal ions for activity (He et al, 1995).

The enzymes were observed to be reversible, depending on the reaction conditions.

Finally, the aforementioned study of gallate decarboxylase in P. agglomerans T71 (Zeida et al, 1998) details the stabilization of the enzyme and sequencing of its amino-terminus.

The authors found that gallate decarboxylase from P. agglomerans T71 is unique among similar decarboxylases in that it requires iron as a cofactor. Information on non-oxidative decarboxylases from microorganisms is accumulating, from both fundamental and applied research, and thus far it can be generalized that these proteins form a class of enzymes which are fairly substrate specific, unstable, and for the most part function independently of cofactors.

1.4 Metabolic engineering for the production of chemicals from

renewable resources

Vast amounts of aromatic carboxylic acids are available as natural products from plant and wood residues. However, exploitation of such substances as starting material for chemical syntheses or as fermentation substrates has not attracted the sustained attention of molecular biologists. Much remains to be learned about the microbiological systems that offer potential avenues for recovering the major biochemical resource that the plant-

derived aromatic carboxylic acids represent. To my knowledge, there are no published

studies of metabolically engineered organisms for the biotransformation of lignin

residues. However, E. coli has been modified, in an elaborate genetic engineering 11 scheme, to bioconvert D-glucose to adipic acid, the building block of nylon (Draths &

Frost, 1994). Briefly, E. coli AB2834, a mutant lacking shikimate dehydrogenase, was transformed with the plasmids pKD136 (encoding a transketolase, DAHP synthase, and

DHQ synthase), pKD8.243A (encoding DHS dehydratase and protocatechuate decarboxylase), and pKD8.292 (encoding catechol 1,2-dioxygenase). The bacterial strain was kept stable due to plasmid compatibility and the use of drug resistance selections for each plasmid. E. coli AB2834/pKD136/pKD8.243A/pKD8.292 utilized its suite of aromatic amino acid biosynthetic genes as well as aromatic acid catabolic genes to effectively convert D-glucose, derived from many agricultural sources, to adipic acid.

The process provides an effective alternative to conventional adipic acid synthesis from benzene, a process that results in the production of high amounts of nitric oxide, the

"greenhouse" gas implicated in the depletion of the ozone layer of the Earth. It should be noted that almost all the genes used in the engineered E. coli strain have been patented, and the DNA sequences of these genes, if known, are not freely available to the public.

Vanillate decarboxylase from Streptomyces sp. D7 could be useful in the future as a critical link in a multi-step metabolic engineering regime similar to that described for E. coli AB2834. The utility of vanillate decarboxylase in value-added biomass conversions will be described in more detail in the Discussion section of this report.

1.5 Streptomyces: versatile soil microbes

Streptomyces is a genus of gram positive, filamentous, sporulating bacteria that are

mainly native to soil, but are also found in aquatic environments. Being predominantly 12 soil-borne microbes (Atlas & Bartha, 1993), these organisms have developed diverse metabolic capabilities that allow for the production of a diverse array of chemical compounds. These secondary metabolites (produced during the late (stationary) phases of growth) have found many applications for humans, ranging from anti-inflammatory therapeutics to potent anti-microbial agents. The high industrial value of certain streptomycete strains for their secondary metabolites has resulted in the development of a detailed understanding of the genetics and physiology of these microorganisms

(Hopwood et al, 1985). Starting from spores, streptomycetes undergo germination to form substrate mycelial (hyphal) growth. Subsequently, the microorganism shifts to a secondary phase of growth, in which aerial mycelia are formed on the substrate mycelial base. It is during this phase that most of the industrially useful compounds are formed.

Upon the induction of signaling responses (such as two-component serine-threonine phosphorylation cascades) due to environmental stimuli such as starvation, the organisms shift to a sporulation cycle, in which chains of uni-nucleate spores are produced at the end of the aerial mycelial tips. The streptomycete growth cycle is illustrated in Figure 4.

At a genetic level, streptomycetes possess linear chromosomes, and may contain both circular and linear plasmids. As members of the Actinomycetes, they characteristically possess DNA with a high guanosine and cytosine (G+C). content, typically in the range of

70-75%. Currently, the Streptomyces coelicolor A3(2) genome sequencing project is well underway, and thus far, of the 8 Mb, >7000 gene (estimated) genome, just over 3

Mb have been sequenced. The data accumulated provides the following statistics (D.

Hopwood, J. Davies, personal communication). The average G+C content is 71.72%. Of the open reading frames (ORFs), 3.5% have been previously sequenced, 50.5% resemble 13 those of known function, 19.2% resemble those of unknown function (hypothetical proteins), and 26.8% have no database match. There is an average of 1.14 kb per ORF, suggesting that the genome is tightly packed ~ by comparison, the yeast Saccharomyces cerevisiae has 13 Mb and <6000 genes, with an average of 1.2 kb per ORF.

Streptomycetes are renowned for their secondary metabolite production, but their abundance in soil, particularly in environments rich in humic matter and lignocellulose, have adapted them to degrade a wide variety of natural substances. These microorganisms provide a major contribution to the global carbon cycle by assisting fungi in the mineralization of cellulose and lignin. Surprisingly, this aspect of

streptomycete biology has been little investigated; Streptomyces viridosporus T7A, which secretes a powerful lignin peroxidase is, thus far, the most well characterized

lignin-degrading streptomycete (Thomas & Crawford, 1998). However, most

streptomycetes isolated from soils do not degrade lignin, but are very efficient at the catabolism of lignin-related, low molecular weight aromatic compounds such as vanillic

acid. A possible scenario (Kirk, 1987) is that in the natural consortia of microorganisms present in forest soils, fungi such as the basidiomycetes perform depolymerization of

intact lignin, leading to the release of more soluble fragments which can be transformed

and mineralized by other microorganisms such as streptomycetes (Figure 5). The non-

oxidative decarboxylation of vanillic acid to guaiacol is one such transformation. Figure 4: The Streptomyces growth cycle. 15

It is the intent of these studies to shed light on the process of non-oxidative decarboxylation of aromatic acids by utilizing a proteomics and functional genomics approach. By studying gene expression patterns during microbial catabolic processes, I have been able to partially characterize the molecular basis by which non-oxidative decarboxylation occurs in Streptomyces, and perhaps in other organisms with similar functions. The following information provides background details regarding the isolation of Streptomyces sp. D7, catabolic phenotyping, proteomic analysis of gene expression, and isolation and partial sequencing of a putative vanillate catabolic enzyme, leading to the work described in this thesis (Chow, 1996).

1.6 Isolation and identification of Streptomyces sp. D7

The organism used for this study, Streptomyces sp. D7, was isolated from a soil sample taken from forest land on the University of British Columbia campus in Vancouver, B.C.,

Canada (Chow, 1996). The organism produces abundant gray spores when grown on a mannitol soya agar medium, and produces a bright yellow diffusable, water-soluble pigment during growth in various solid and liquid media.

Figure 5 (following page): A possible niche for Streptomyces in lignin degradation. 16

Fungal lignin solubilization 17

1.7 Catabolic tests and 2D-PAGE analysis

Streptomyces sp. D7 was determined, by UV spectrophotometry and HPLC analyses of culture supernatants, to efficiently bioconvert vanillate to guaiacol (Figure 6), suggestive of the activity of a non-oxidative decarboxylase (Chow et al, 1999). While Streptomyces sp. D7 was apparently capable of limited growth using the carboxylic acid moiety of vanillate as a sole carbon source, no further degradation of guaiacol was observed. As mentioned previously, other microorganisms and strains of Streptomyces have been shown to perform this enzymatic reaction (Crawford & Olson, 1978; Pometto III et al,

1981; Sutherland et al, 1981), but thus far, no genetic information has been published regarding these enzyme systems. In order to identify proteins synthesized during vanillate catabolism, high-resolution 2D-PAGE technology (the "Investigator System",

Genomic Solutions Inc.) was used to visualize "genetic snapshots" of cellular activity when growing cells of Streptomyces sp. D7 were induced with non-inhibitory amounts of vanillate (Chow et al, 1999; Chow, 1996). Streptomyces sp. D7 was grown in a mineral salts medium supplemented with 0.5% yeast extract until mycelia were in early logarithmic growth phase. At this point in the growth cycle, the culture was divided and

3.6 mM vanillic acid was added to one of the cultures to induce a response. Aliquots of both induced and uninduced cultures were pulse labeled with 33S-methionine/cysteine at

1, 2, 5, 12, and 15 hours post-induction. A diagram depicting this experiment is shown in

Figure 7. The labeled mycelium samples were sonicated to extract total cell protein and separated by 2D-PAGE. Compilation and analysis, by PDQUEST software (PDI, Inc.) using a SparcStation 5 workstation (Sun Microsystems), of numerous 2D-PAGE gels from several time course experiments resulted in the identification of at least six major 18 proteins that were synthesized in response to vanillate. The most prominent and abundant protein, of 52 kDa in molecular mass with a pi of 4.9 (Figure 8), was pooled from replicate gels, blotted to PVDF membrane and Edman-degradation sequenced, yielding sufficient and reliable amino-terminal data (AYDDLRYFLDTLEKEGQLLRIT) to allow synthesis of a degenerate oligonucleotide probe. This probe allowed me to proceed with the cloning of the gene encoding the 52 kDa, vanillate-induced protein in order to isolate and characterize the genetic elements forming the basis for vanillate decarboxylation in this organism. Such experiments form the basis for this thesis and are described herein. 19

3.5 _ •

3.0

0 10 20 30 40 50 60 Time (h)

Figure 6: Decarboxylation of vanillic acid to guaiacol by Streptomyces sp. D7. Concentration of aromatic compounds in culture supernatants was measured using HPLC and known concentration standard solutions. Vanillic acid concentration is represented by squares (• ); guaiacol concentration is represented by circles (•). Cells were grown to late log phase in YEME liquid medium, harvested, washed and resuspended in MSMYE minimal medium with approximately 3.6 mM vanillic acid. Each time point represents a concentration as measured by HPLC of culture supernatant samples. 20

t +12 t +18 hours +5 hours hours

Cell Density

Substrate addition +2 hours (induction) +1 hour 0 (pre-induction)

Time

Figure 7: Experimental scheme for growth and vanillic acid induction by Streptomyces sp. D7 cells for 2D-PAGE analyses. 1 ml aliquots of mycelia were radioactively pulse labeled at the times indicated. The panels in Figure 6 (following page) correspond to these time points. (Chow, 1996) 21

t = 1 h t = 2h t = 5h t = 12.5h t=18h

Figure 8: Protein 2D-PAGE profile of the synthesis of protein 3717 by Streptomyces sp. D7 in response to 3.6 mM vanillic acid. 2D-PAGE profiles of an uninduced culture are also shown for comparison. Time values are measured as hours post induction. Arrows highlight the area in which protein 3717 appears. The panels are enlargements of the 52 kDa, pi 4.9 region from eleven different 2D-PAGE gels representing eleven time point samples and two treatments, (from Chow et al, 1999) 22

1.8 Objectives

The objectives for this study are as follows:

1. Clone and sequence the gene encoding the Streptomyces sp. D7 52 kDa protein

induced by vanillic acid (the putative vanillate decarboxylase gene).

2. Sequence regions up- and downstream of the gene encoding the 52 kDa protein, and

locate other open reading frames in the immediate vicinity.

3. Compare the putative vanillate decarboxylase gene and its translation product to

similar genes and proteins from other microorganisms.

4. Determine if the putative vanillate decarboxylase gene is present in other

streptomycetes.

5. Confirm the function of the putative vanillate decarboxylase gene by obtaining a

gene knockout mutant of Streptomyces sp. D7, expressing the cloned gene in S.

lividans 1326 as a recombinant host, or by expression of the gene in E. coli.

6. Characterize purified vanillate decarboxylase enzyme, if sufficient quantities can be

obtained. 23

2. MATERIALS AND METHODS

2.1 Bacterial strains and plasmids

2.1.1 Bacterial strains

Bacterial strains used in this study are shown in Table 1. Streptomyces sp. D7 was isolated from forest soil on the University of British Columbia campus, as described above. Streptomyces lividans 1326, which was used for gene expression experiments, was obtained from the John Innes Collection in Norwich, United Kingdom. Escherichia coli DH5a was obtained from Gibco BRL and used for general DNA cloning and sequencing procedures. Escherichia coli BL21(DE3), which served as host for the T7

RNA polymerase gene expression system, was obtained from Novagen as a component of the pET22b(+) gene expression kit.

2.1.2 Plasmids - cloning and isolation

Plasmids used in this study are listed in Table 1. Subcloning of Streptomyces DNA (to be described in detail in Section 2.3) was performed in Escherichia coli DH5a with pUC19. Transformation of E. coli DH5a was achieved using a heat shock protocol, in which plasmid DNA was incubated with competent cells for 30 minutes on ice, then placed at 37°C for 30 seconds, then another 2 minutes on ice. One milliliter of SOC liquid medium (20 g of Bacto-Tryptone, 5 g of yeast extract, 0.5 g of NaCI, 10 ml of 250

mM KC1, 950 ml of distilled water, pH to 7.0; before use, add 5 ml of 2 M MgCl2 and 20 ml/L of sterile 1 M glucose) was added to the transformation mixture followed by one hour incubation at 37°C. Aliquots were plated on LB agar supplemented with an 24 antibiotic appropriate for selection of the plasmid being transformed (for example, 100

ug/ml ampicillin for pUC19).

Expression studies of the VDC genes in Streptomyces lividans 1326 (Section 2.8) were performed using pIJ680 (Figure 9) (Hopwood et al, 1985), a vector that places target genes under the control of the aminoglycoside phosphotransferase (aph) constitutive promoter, and pIJ702 (Figure 10) (Katz et al, 1983), which provides the weaker constitutive tyrosinase (mei) promoter. S. lividans 1326 was converted to protoplasts and transformed according to published methods (Bibb et al, 1978; Thompson et al, 1982).

Protoplasts were plated on R5 solid medium (Thompson et al, 1980) and allowed to regenerate for 14 hours before transformants were selected by an overlay of soft nutrient agar containing thiostrepton to achieve a final concentration of 50 pg ml"1 thiostrepton per plate.

Chromosomal DNA was extracted from Streptomyces strains by the method of Fisher

(Hopwood et al, 1985). Streptomyces plasmids were isolated by an alkaline lysis procedure (Hopwood et al, 1985) and E. coli plasmid DNA was routinely isolated using the Qiaprep Spin miniprep kit (Qiagen) or the NucleoSpin miniprep kit (Clontech) for sequencing and routine manipulations.

Table 1 (following 2 pages): Bacterial strains and plasmids used in this study. Construction of gene expression plasmids is described in detail in a subsequent section of Materials and Methods. 25

Strain or plasmid Relevant properties Reference/source E. coli DH5a MCR Host for pUC19and Gibco BRL derivatives

Streptomyces Streptomyces sp. D7 Wild-type vanillate Chow, 1996 decarboxylase isolate

Streptomyces lividans 1326 Wild-type Streptomyces John Innes Collection, heterologous expression Norwich host

Plasmid vectors pUC19 2.7 kb Apr E. coli cloning Gibco BRL vector pKCEl pUC19 carrying 4.4 kb This study; Chow et al., BamHI sub-fragment of the 1999 ~13 kb Streptomyces sp. D7 genomic DNA piece cloned from the phage library; contains vdcB, vdcC, vdcD pKCE2 pUC19 carrying 527 bp Sail This study sub-fragment of pKCEl insert (see above) pKCE3 pUC19 - 3 kb BamHI sub- This study fragment of phage clone pKCE4 pUC19-2.2kb£amHI This study sub-fragment of phage clone pKCE5 pUC19- 1.9 kb BamHI This study sub-fragment of phage clone pKCE6 pUC19 - 0.8 kb BamHI This study sub-fragment of phage clone

( 26

Table 1 (continued): Strain or plasmid Relevant properties Reference/source pIJ702 7.2 kb Tsr Streptomyces Katzetal., 1983/ cloning vector with mei TerraGen Discovery, Inc. promoter pIJ680 5.3 kb Tsr Streptomyces Hopwood et al, 1985/ expression vector with aph Dr. L. Sandercock, UBC promoter Biotechnology Laboratory pKCSl pIJ702 carrying 4.4 kb This study; Chow et al., BamHl insert from pKCEl 1999 inserted in same orientation as mei promoter pKCS2 pIJ702 carrying 4.4 kb This study; Chow et al, BamHl insert from pKCEl 1999 inserted in opposite orientation as mei promoter pKCS4 pIJ680 carrying PCR This study amplified vdcB inserted downstream of aph promoter pKCS5 pIJ680 carrying PCR This study amplified vdcC inserted downstream of aph promoter pKCS6 pIJ680 carrying PCR This study amplified vdcD inserted downstream of aph promoter pKCS7 pIJ680 carrying PCR This study amplified vdcBC inserted downstream of aph promoter 27

Table 1 (continued): Strain or plasmid Relevant properties Reference/source pKCS8 pIJ680 carrying PCR This study amplified vdcCD inserted downstream of aph promoter pKCS3 pIJ680 carrying PCR This study; Chow et al., amplified vdcBCD inserted 1999 downstream of aph promoter 28

2.1.3 16s rDNA sequence-based strain identification

Streptomyces sp. D7 was characterized by sequencing a 505 base pair 16S ribosomal

DNA fragment produced using streptomycete specific PCR primers. The primers consisted of the following sequences: forward: 5'-GAGATTTGATCCTGGCTCAG-3'; reverse: 5'-CGGACTGGTTGTTACGACTTC-3'. Thermocycling was performed as follows: 1 minute denaturation at 95°C, 2 minutes annealing at 55°C and 2 minutes extension at 72°C. The cycle was repeated 30 times, with a final extension of 10 minutes at 72°C.

2.2 Media and growth conditions

Streptomyces sp. D7 and S. lividans 1326 were routinely cultivated in tryptic soy broth

(TSB) or on mannitol soy flour agar plates at 30°C. Catabolic tests and growth experiments were performed using mineral salts medium supplemented with 0.5% yeast

1 1 1 extract (MSMYE: (NH4)2S04 0.1 g L" , NaCI 0.1 g L' , MgS04-7H20 0.2 g L" , CaCl2

1 1 1 1 0.01 g L" , yeast extract 0.5 g L" , K2HP04 1.0 g L" , KH2P04 0.5 g L" , pH 7.2) and aromatic compounds of interest at concentrations of 3.6 mM to 6 mM. When appropriate, thiostrepton at 50 (j,g ml"1 was included for selection and maintenance of plasmid containing strains. For DNA extraction or protoplast preparation, strains were

cultivated in YEME (liquid medium) supplemented with 0.5% glycine and 5 mM MgCl2 at 30°C (Hopwood et al, 1985). E. coli DH5oc was grown in Luria-Bertani (LB) medium

(supplemented with 100 \xg ampicillin ml"1), when maintaining pUC-based plasmids) at

37°C. Figure 9: Map of expression vector pIJ680 (Hopwood et al, 1985) 30 31

2.3 Library construction and gene cloning

A Lambda DASH II (Stratagene) genomic DNA phage library of chromosomal

Streptomyces sp. D7 fragments was constructed by ligating 9-22 kb Sau3AI partially digested chromosomal DNA fragments into the BamHl site of the phage arms. Phage carrying genomic DNA fragments were mixed, in soft nutrient agar, with Escherichia coli XL 1-Blue MRA(P2) cells and plated on NZY agar plates. The plates were incubated at 37°C overnight and the resulting plaques (approximately 4000 - between 300 and 400 plaques per plate) were lifted with Hybond-N nylon membranes (Amersham). The library was screened by hybridization at 60°C with a y32P-ATP-labeled 56-mer oligonucleotide probe 3717C (5'- GC(CG) TAC GAC GAC CT(CG) CG(CG) TAC TTC

CT(CG) GAC AC(CG) CT(CG) GAG AAG GAG GG(CG) CAG CT(CG) CT -3') derived from protein amino-terminal sequencing data. Of the approximately 4000 plaques in the library, twelve hybridized strongly to the probe. Lambda phage from one of these plaques were isolated and propagated in XL 1-Blue MRA(P2) E. coli for DNA isolation (see Section 2.3.1). The vdcB, vdcC and vdcD genes were subcloned on a 4.4 kb BamHl fragment into pUC19 in Escherichia coli DH5oc MCR (Gibco BRL).

2.3.1 Lambda phage DNA preparation

Lambda DASH II phage DNA carrying Streptomyces sp. D7 chromosomal DNA BamHl

fragments was isolated by the following protocol. Ten milliliters of LB liquid medium

was inoculated with one agar plug (from the NZY/soft agar/phage plates (described

above) containing a phage plaque), 50 uL of XL 1-Blue MRA(P2) E. coli (in 0.01 M

MgS04, O.D.6Q0 -0.5), 100 uL of 1 M MgS04, and shaken at 37°C overnight. After the 32 incubation period, 100 uL of chloroform was added and the culture was shaken for an additional 2 minutes at 37°C, then centrifuged at room temperature for 10 minutes at

5000 x g. The aqueous phase was saved, and the following were added: 100 uL 1 M

MgS04, 10 ml TM buffer (50 mM Tris-HCl pH 7.4, 10 mM MgS04), 32 iaL 10 mg/ml

DNAse and 10 mg/ml RNAse. The solution was incubated at room temperature for 15 minutes, then 2 ml of 5 M NaCI and 2.2 g of PEG (m.w. 6000-8000) were added and allowed to dissolve completely. The mixture was the incubated for 15 minutes on ice, then centrifuged for 10 minutes at 4°C at 10000 x g. The supernatant was discarded, leaving the phage pellet, which was resuspended in a minimum of 300 uL of TM buffer.

The suspension was transferred to a 1.8 ml Eppendorf tube, 300 uL of chloroform was added, mixed, then the tube was centrifuged for 5 minutes at 12000 x g. The aqueous phase was transferred to a new Eppendorf tube, and the chloroform extraction step was repeated until no interface was observed between the chloroform and aqueous phases.

Fifteen microliters of 0.5 M EDTA, 30 uL of 5 M NaCI, and 350 uL of tris-buffered phenol was added, vortexed, then centrifuged for 5 minutes at 12000 x g and the aqueous phase was saved. One final chloroform extraction (350 uL) was performed, then 875 uL of 100% cold ethanol was added and incubated at -20°C overnight to precipitate the

DNA. After centrifugation at 12000 x g, the ethanol was removed, the DNA pellet was

air dried briefly, and resuspended in 50 to 100 uL of TE buffer.

2.4 DNA sequencing and analysis.

Automated DNA sequencing was performed using the AmpliTaq PRISM kit (Applied

Biosystems) with a standard thermocycling program provided by the manufacturer, with 33 variations in the annealing temperature to match the melting temperature of the sequencing primer being used. Sequencing reactions were carried out by the Nucleic

Acid Protein Sequencing (NAPS) Unit at the University of British Columbia and electrophoresed on an ABI Model 377 DNA sequencing apparatus (Applied Biosystems).

Nucleic acid sequence was analyzed by the Wisconsin Package Version 10 (Genetics

Computer Group) on a Sun Microsystems SparcStation5 (Sun Microsystems).

2.5 Southern blotting and hybridization

Genomic DNA preparations from Streptomyces sp. D7 and Streptomyces setonii 75Vi2 were digested with the restriction endonuclease Sail overnight at 37°C and electrophoresed in a 0.7% agarose gel. The gel was photographed, soaked in 0.15N HC1 for 15 minutes for depurination (to facilitate transfer of high molecular weight fragments) and soaked in alkaline hybridization solution (0.6M NaCl, 0.4M NaOH) for an additional

15 minutes. The treated gel was then Southern blotted to a positively charged nylon membrane (Boehringer Mannheim) using the same alkaline solution as a transfer buffer.

After disassembly of the transfer apparatus, the membrane was rinsed briefly in 2X

sodium saline citrate (SSC) and baked at 80°C for one hour. Hybridizations utilized a rotating incubation chamber (Hybaid) and glass hybridization bottles. The membrane

was soaked in 2X SSC, rolled, placed in a glass hybridization bottle with hybridization

solution (5X Denhardt's Solution, 6X SSPE, 0.1% SDS) and prehybridized at 65°C for

several hours. After prehybridization, the P-labeled probe (a- P or y- P, depending on

the labeling procedure) was denatured at 95 °C for 5 minutes, cooled on ice briefly,

centrifuged, then added to the hybridization bottle containing the membrane. 34

Hybridization was carried out overnight at 65°C. The wash regimen consisted of 2 washes of 2X SSC, 0.1% SDS for 10 minutes at room temperature, IX SSC, 0.1% SDS for 15 minutes at 65°C once, then a brief room temperature rinse in 0.1X SSC, 0.1%

SDS. The washed blot was semi-dried, then placed in plastic wrap and exposed to film

(Kodak X-AR) for several hours to overnight. An identical hybridization procedure was used during the phage library screening process, and also subsequent cloning manipulations.

2.6 Chemical analyses

Culture supernatants were filtered through 0.45 urn syringe filters and processed through a C-18 hydrophobic interaction column attached to a HPLC system (Hewlett Packard,

Model 1050). Conditions for separation were 30% phosphoric acid/water, 70% methanol, with a flow rate of 1.0 ml min"1. Retention times for vanillic acid and guaiacol under these conditions are 5 minutes and 4 minutes, respectively. Integrated peak areas corresponding to compounds in supernatant samples were calibrated against known concentrations of vanillic acid and guaiacol standards. Additional analysis of supernatant samples was performed using a Cary 1 Bio ultraviolet/visible spectrophotometer

(Varian). For UV analysis, vanillic acid characteristically displays a primary absorbance at 250 nm and a secondary absorbance at 285 nm, while guaiacol absorbs at 275 nm.

2.7 RNA isolation and analysis

Total RNA was isolated from cells grown under two different sets of conditions. Primary

cultures were grown in 25 ml YEME for 48 hours before cells were pelleted by 35 centrifugation, washed twice with sterile water and resuspended in minimal media

(MSMYE). For induced cultures, the media was supplemented with 3.6 mM vanillic acid, while no additional substrates were added to the uninduced control. These cultures were then allowed to grow an additional three hours after which cells were pelleted and washed as before. Total RNA was isolated using standard RNA isolation techniques

(Hopwood et al, 1985; Kirby et al, 1967). For transcript detection, Northern gel electrophoresis and transfer were performed according to the manufacturer's recommendations for the NorthernMax kit (Ambion). 15 ug total RNA was loaded per lane in a polyacrylamide gel and 1 ug of an RNA standard ladder (NEB) was included for size comparison. After electrophoresis was complete, the RNA ladder lane was excised and stained with ethidium bromide to allow for visualization and to confirm RNA integrity. Transcript was detected using a probe specific for the vdcC gene. To generate the probe, traditional double-stranded PCR was performed on pKCEl using oligonucleotides vdcC.F and vdcC.R (Table 2). Following amplification, excess dNTPs and oligonucleotides were removed using a QIAquick PCR Purification Kit (Qiagen).

This product was then used as template for asymmetric PCR with only the vdcC.R primer. This resulted in a single-stranded PCR product that was complementary to the predicted RNA transcript. During chain elongation, j2P-dCTP was provided in place of dCTP in the dNTP mix to allow for direct incorporation of radiolabel. The extension product was purified and allowed to hybridize with the immobilized RNA at 60°C for 24 hours. Excess probe was removed by washing as directed by the NorthernMax protocol, and the hybridizing transcript was visualized by exposure to autoradiography film

(Kodak XAR) for exposure and development. 36

Table 2 (including next page): Polymerase chain reaction (PCR) primers used in this study. Engineered restriction enzyme sites are underlined. Primer Length Sequence (5' —» 3') Purpose (nt) Hybridiza• tion probes vdcB.F 26 ACAGGTCAGCGACAGG Forward amplification of vdcB gene TTTGAGGTGG DNA. vdcB.R 21 TACGGGGCAGGGGACT Reverse amplification of vdcB gene TCAGG DNA. vdcC.F 20 GGCGACGCCGCCTGAA Forward amplification of vdcC GTCC gene DNA. vdcC.R 20 GGGTCGGTCGGTGTCA Reverse amplification of vdcC gene GACG DNA. vdcD.F 20 CACCGATCCTCACTGA Forward amplification of vdcD AAGG gene DNA. vdcD.R 20 CATAGACCGCGTGCCG Reverse amplification of vdcD gene GTCG DNA. pIJ680 expression vdcBCD.FX 26 CGGATCCAGTGACAGG Forward amplification of vdcB, TTTGAGGTGG vdcBC and vdcBCD genes for expression in pIJ680. Engineered BamHI site. vdcBCD.RX 28 AGTCTAGACCGGCGTC Reverse amplification of vdcD, GGAGGGATGACC vdcCD and vdcBCD genes for expression in pIJ680. Engineered Xbal site. vdcB.RX 21 TTCTAGACAGGGGACT Reverse amplification of vdcB gene TCAGG for expression in pIJ680. Engineered Xbal site. vdcC.FX 29 CGGATCCCCCGTAAAG Forward amplification for vdcC GAATTCACCATGG gene and vdcCD genes for expression in pIJ680. Engineered BamHI site. 37

vdcBC.RX 29 TTCTAGATCAGACGCG Reverse amplification for vdcB, GGCCGCGATCAGG vdcC and vdcBC genes for expression in pIJ680. Engineered Xbal site. vdcD.FX 28 CACGGATCCTCACTGA Forward amplification of vdcD AAGGACAACTCC gene for expression in pIJ680. Engineered BamHl site. pET22b(+) expression vdcB.pETF 21 TCATATGCGGTTGGTCG Forward amplification of vdcB gene TGGG for expression in pET22b(+). Engineered Ndel site. vdcB.pETR 21 TTCTCGAGAGGGGACT Reverse amplification of vdcB gene TCAGG for expression in pET22b(+). Engineered Xhol site. vdcC.pETF 23 GGAATTCCATATGGCCT Forward amplification of vdcC ATGACG gene for expression in pET22b(+). Engineered Ndel site. vdcC.pETR 18 CGCGGGCTCGAGTCAG Reverse amplification of vdcC gene GC for expression in pET22b(+). Engineered Xhol site. 38

2.8 Gene expression

2.8.1 Cloning and expression in pET22b(+) in Escherichia coli BL21

DNA encoding the vdcB and vdcC genes was PCR-amplified, using primers listed in

Table 2, to incorporate Ndel and Xhol sites upstream and downstream, respectively of the target gene(s) prior to cloning into the pET22b(+) vector (Novagen). These plasmid constructs were transformed into E. coli BL21(DE3), which were grown to OD 0.4, then induced with IPTG to activate expression by the T7 polymerase system, characteristic of the pET22b(+) vector. Expression was performed for two hours, at which time the cells were pelleted and protein extracted. To optimize protein expression, induction of the T7 system was performed at 37°C, 30°C and 25°C, using 0.1 mM or 1 mM IPTG.

Reductions in temperature and IPTG concentration are commonly used methods to increase the chances of obtaining properly folded proteins.

2.8.2 Cloning and expression in pIJ702 in S. lividans 1326

The 4.4 kb BamHI DNA fragment containing the vdcBCD gene cluster was inserted into the Streptomyces cloning vector pIJ702 at the unique Bglil site. Insertion at this site places the cluster downstream of the mel promoter, thereby disrupting transcription of the tyrosinase gene that serves as a color selection marker for transformants. pIJ702 carrying the insert in the same orientation as the mel promoter was designated pKCSl; conversely, a vector construct with the insert in the opposite orientation to the promoter was designated pKCS2 (Figure 11). 39

Figure 11: pIJ702 and the 4 kb BamHl Streptomyces sp. D7 genomic DNA fragment inserted in both orientations in the tyrosinase gene (mei) promoter. 40

2.8.3 Cloning and expression in pIJ680 in S. lividans 1326

In order to identify the gene sequences encoding the decarboxylase, DNA regions were

PCR amplified using specific primers (Table 2) which included a BamHI site upstream, and a Xbal site downstream, of the gene(s). The PCR-generated genes were cloned downstream of the aph promoter in BamHl-Xbal cut pIJ680, replacing most of the aph gene. The following genes were amplified using this BamHI-Xbal PCR cloning strategy for ligation into pIJ680, creating new plasmids designated in brackets: vdcB (pKCS4), vdcC (pKCS5), vdcD (pKCS6), vdcBC (pKCS7), vdcCD (pKCS8), vdcBCD (pKCS3). pIJ680 and the PCR-generated inserts are shown in Figure 12. All plasmids were transformed individually into 5*. lividans 1326 and the resulting recombinant hosts were screened for the ability to decarboxylate vanillic acid. fl m CO r- oo ro co co in co co co o o o o o o X. X. X. ^ X. a a a aa a

Figure 12: pIJ680 and the PCR-generated gene combinations inserted downstream of the promoter for the aminoglycoside phosphotransferase (aph) gene. 42

2.9 Enzyme assays

Late log phase mycelia harvested from YEME cultures were washed with phosphate buffer (10 mM Tris pH 8.0, 1 mM EDTA, 1 mM DTT) and resuspended in the same buffer, containing protease inhibitors (100 jag ml"1 PMSF, 1 ug ml"1 Pepstatin A), at a ratio of 0.1. ml buffer per 1 ml of original culture. Samples were sonicated as previously described and centrifuged at 12,000 x g for 15 minutes to remove insoluble cell debris.

Soluble cell extracts were tested for decarboxylase activity by adding vanillic acid or comparative substrates to a final concentration of 1 mM. Samples were incubated for 15 minutes at 25°C, at which time they were analyzed by scanning the UV range from 300 nm to 200 nm. Soluble cell extract without substrate was used as a background in the reference cuvette. To test enzyme activity under anaerobic conditions, nitrogen gas was slowly bubbled through assay sample tubes prior to addition of substrate.

2.10 Recombinant protein purification from E. coli BL21(DE3)

As mentioned in Section 2.8.1, vdcB and vdcC were recombinantly expressed using the pET22b(+) plasmid vector in E. coli BL21(DE3). Attempts were made to purify the proteins produced by these strains. Under all expression conditions, inclusion body material was obtained, and thus efforts were mainly focused on extracting and refolding the insoluble protein aggregates. However, in the event that minute amounts of soluble, active enzyme were produced, cell lysates were passed through nickel-based affinity columns to purify the histidine-tagged recombinant proteins. 43

2.10.1 Lysate preparation

Ten milliliters of LB medium containing 100 pg/mL ampicillin were inoculated with material from one colony of the appropriate E. coli BL21(DE3) expression host and grown overnight at 37°C with shaking. Subsequently, 50 ml of prewarmed media containing 100 pg ml"1 ampicillin was inoculated with 2.5 ml of the overnight culture and grown at 37°C, with shaking, until the OD600 reached approximately 0.6 (about one hour). IPTG was added to a final concentration of either 1 mM or 0.1 mM, and cultures were grown for an additional 4-5 hours at 37°C, 30°C, or 25°C, depending on the desired expression conditions. As has already been mentioned, reduction of IPTG concentration and growth temperature is believed to reduce the amount of inclusion body formation due a slowing of the expression process. After the expression period, cells were harvested by centrifugation at 4000 x g for 20 minutes. The cell pellet was resuspended in 5 mL of

lysis buffer (50 mM NaH2P04, pH 8.0; 300 mM NaCl; 10 mM imidazole) and subjected to a freeze/thaw cycle using dry ice/ethanol and cold water, repeated three times. The

sample was then sonicated with a microtip 6 times, on ice, at 10 seconds per burst with

10 second pauses, at 200-300 watts. The lysate was then centrifuged at 10000 x g at 4°C for 30 minutes. At this point, the supernatant contained any soluble recombinant protein, while the pellet contained the large mass of inclusion body material, visible as a white

layer on top of the light yellow cell debris.

2.10.2 Inclusion body protein purification, solubilization and refolding

The pellet was resuspended in 0.1 culture volume of inclusion body wash buffer (200

mM Tris-HCI, pH 7.5, 100 mM EDTA, 10% Triton X-100) and recentrifuged. This wash 44 step was repeated, and the resulting purified inclusion bodies (of which the wet weight was noted) were solubilized in inclusion body solubilization buffer (500 mM CAPS, pH

11.0, 0.3% N-laurosarcosine) at 10-20 mg ml"1 for 15 minutes at room temperature. The

(mostly) solubilized solution was centrifuged at 10000 x g for 10 minutes at room temperature, and the supernatant was removed to a clean tube. The soluble protein solution was placed in dialysis tubing, and the sample was dialyzed against 20 mM Tris-

HCl, pH 8.5, 0.1 mM DTT, using at least three buffer changes of greater than 50 times the sample volume. Each dialysis step was performed for at least 3 hours at 4°C. After dialysis, the sample was concentrated using a Centricon-10, 10000 m.w. cut-off ultrafiltration device and analyzed by SDS-PAGE for purity.

2.10.3 Soluble protein purification procedure

The soluble fraction of the recombinant E. coli BL21(DE3) cell lysates were processed through nickel-nitrilotriacetic acid (Ni-NTA) metal-affinity chromatography matrix columns in an attempt to purify any proteins that were not produced as inclusion bodies.

Ni-NTA mini spin columns (Qiagen, catalogue #31313) were used for small-scale experiments, while nickel resin (ProBond, Invitrogen, catalogue #R801-01) was packed in a larger column for scaled-up procedures. Column loading, washing and elution were performed according to the manufacturers' instructions.

2.11 Construction of a knockout allele of the vdcC gene

To elucidate the function of vdcC, the gene encoding the 52 kDa protein, a disrupted

allele was created by removing a portion of the 5' region of the gene and replacing it with 45 the thiostrepton resistance gene (tsr). This strategy utilized pSLl 190 (Pharmacia), which contains a superlinker with 78 restriction enzyme sites, allowing for more versatility in the construction process. The allele was designed for gene knockout experiments, which are described in subsequent sections. The allele construction strategy, using pIJ702, pSLl 190 and pUC19, is shown in Figure 13.

Figure 13: Strategy for a knockout of the vdcC gene using the thiostrepton (tsr) resistance gene, (a) Preparation of the thiostrepton resistance marker cassette; (b) Insertion of the thiostrepton resistance cassette in vdcC to create the knockout allele; (c) protoplast transformation and double crossover event. 46

8i| M (a) Zmtp

Bell

Bell BamHI, Bglll

S IS "9

Bcll^ tsr Bell I * 1 1,1 kb

ligation

Dral

Ncol

Ncol,

Dral Ncol (sft Dral(blunt)

NOTE: tsr1" gene orientation could be in either direction. 47

KNOCKOUT ALLELE 4«

PLATE PROTOPLASTS ON REGENERATIVE MEDIA, SELECT FOR THIOSTREPTON RESISTANCE 49

2.12 Streptomyces sp. D7 protoplast preparation and transformation

Streptomyces sp. D7 protoplasts were prepared and transformed as described for

Streptomyces lividans 1326 in Section 2.1.2, with the exception that R6 agar plates were used instead of R5. The regeneration time for Streptomyces sp. D7 was several days slower than that observed for S. lividans 1326.

The recipe for R6 solid regeneration agar medium is as follows (grams per liter): mix and autoclave 200 g sucrose, 10 g dextrose, 1 g casamino acids, 0.05 g MgSO^H^O, 0.1 g

K2SO4, 20 g agar. After autoclaving, add 11 g glutamate, 1 mL standard trace elements

solution, 7 g CaCl2-2H20 and 100 mL MOPS buffer (0.1 M, pH 7.2).

2.13 E. coli - Streptomyces interspecies conjugation

To obtain a gene knockout mutant, attempts were made to transform Streptomyces sp. D7 mycelia by conjugation with E. coli. As an evaluation of this method, E. coli SI7-1 carrying the vector pPM801 (aph+, tra+) was mated with Streptomyces sp. D7 mycelia to assess whether the vector, which encodes resistance to the antibiotic kanamycin, was transferred between the species.

2.13.1 E. coli S17-l/pPM801 preparation

One colony of E. coli S17-l/pPM801 was diluted in 5 mL of LB broth, supplemented with 10 pg/mL kanamycin, and incubated for 8 hours at 37°C. The grown culture was diluted 1:10 with fresh LB broth with kanamycin. 50

2.13.2 Streptomyces sp. D7 preparation and mating

Streptomyces sp. D7 spores were scraped from a lawn of bacteria on a mannitol-soy agar plate and resuspended in 5 mL of TES buffer (0.05 M, pH 7.2). The suspension was heat-shocked at 50°C for 10 minutes, then placed under cold tap water to stop the shock process. The spores were then centrifuged at 3000 x g and resuspended in 1 mL of LB broth. 150 uL of the spore suspension was plated on an AS-1 agar plate (Hopwood et al,

1985) and the spores were allowed to grow for 2 to 4 hours at 30°C. Subsequently, 100 pL of the E. coli S17-l/pPM801 preparation (see above) was plated on top of the

Streptomyces sp. D7 spores. After overnight growth of the E. coli layer, the layer was gently scraped off using a sterile glass spreader and 0.05 M TES buffer, pH 7.2. An antibiotic overlay, consisting of 2.5 mL of 50 ug/mL nalidixic acid and 25 jag/mL kanamycin, was spread over each plate. Plates were allowed to incubate for a further 6 days, at which time they were observed for any ex-conjugants.

2.14 Reagents and enzymes

All reagents used were of the highest quality, and purchased from Sigma Chemical

Company unless noted otherwise. Restriction endonucleases and other modification enzymes were obtained from Gibco BRL, New England Biolabs, Boehringer Mannheim or Pharmacia. 51

3. RESULTS

3.1 16s rDNA sequence identification

Sequence of the 505 bp PCR product amplified from Streptomyces sp. D7 16s rDNA was matched against the GenBank database using the BLAST-N program (Altschul et al,

1990). The result was an exact match (data not shown) with the 16s rDNA sequence from Streptomyces sp. strain B71277, an isolate from the BBSRC Institute for Food

Research in Reading, United Kingdom (Hutson & Collins, 1997). Studies, if any, involving Streptomyces sp. strain B71277 remain unpublished at the time of writing this thesis, and attempts to contact the authors of the GenBank submission failed to generate a response. This identification procedure confirmed Streptomyces sp. D7 as a member of the streptomycetales. Previously, the microorganism had been referred to as

Streptomyces violaceusniger, as identified by fatty acid methyl ester analysis (Chow,

M.Sc. thesis, 1996). However, the results of the 16s rDNA analysis, considered a more powerful tool for taxonomy (J. Davies, personal communication), suggest otherwise.

3.1 Cloning of the VDC gene cluster

Edman degradation sequencing of protein 3717, isolated from protein 2D-PAGE gels

(previous work described in Section 1.6), yielded the following N-terminal amino acid sequence: AYDDLRYFLDTLEKEGQLLRIT. This sequence matched well with the amino-terminal sequence of />-hydroxybenzoate carboxy-lyase from the anaerobe,

Clostridium hydroxybenzoicum (He and Wiegel, 1995) as shown in Figure 14. The deduced degenerate oligonucleotide probe 3717C (for sequence, see Materials and 52

Methods) was synthesized and hybridized against a lambda DASH II phage library

(Stratagene) of Streptomyces D7 genomic DNA. A phage clone, designated 3717C(+), hybridized strongly to the probe (Figure 15). DNA from phage clone 3717C(+) was purified and digested with the restriction enzyme BamHI (Figure 16). This restriction digest revealed that the phage clone carried an approximately 13 kb Streptomyces sp. D7 genomic DNA insert. Streptomyces genomic DNA BamHI digestion products of approximately 4.4 kb, 3.0 kb, 2.2 kb, 1.9 kb and <1 kb were subcloned individually into pUC19 for further manipulations and sequencing (Figure 17). The pUC19 subclones were designated pKCEl through pKCE6 (Figure 18). pKCEl contains the 4.4 kb BamHI fragment encoding the vanillate decarboxylase gene cluster, while pKCE2 contains a 527 bp Sail fragment that encodes the amino-terminal region of the vdcC gene.The entire sequence of the 4.4 kb Streptomyces sp. D7 DNA fragment (GenBank accession number

AF134589) is shown in Figure 19.

Sequence analysis revealed that the gene encoding protein 3717 was contained on a 4.4 kb BamHI fragment, and was determined to be the second gene in a cluster of at least three genes, designated vdcB (602 bp), vdcC (1424 bp) and vdcD (239 bp), as depicted in

Figure 20. BLAST-X sequence analyses (Altschul et al, 1990) revealed that the gene cluster, in whole or in part, is present in a variety of microorganisms. The vdcB translation product is highly similar to phenylacrylate decarboxylase (PAD) from

Saccharomyces cerevisiae. The yeast PAD contains a putative trans-membrane domain close to the amino-terminus (annotated in GenBank accession number S62017), which is highly conserved among other hypothetical PAD homologues from various 53 microorganisms, as revealed by genome projects (this region is highlighted in the amino acid sequence alignments shown in Figure 21). The vdcC translation product is also highly similar to hypothetical proteins identified in various microbial genome sequencing projects, in addition to the amino-terminal similarity to 4-hydroxybenzoate carboxy-lyase as revealed from the Edman degradation sequencing of protein 3717. Dendrograms of the vdcB and vdcC translation products in comparison to other microbial homologues are shown in Figure 21. Unlike the first two genes in the cluster, the vdcD translation product shows similarity only to a hypothetical protein from Bacillus subtilis. Although these genes have homologues in a number of microbial genomes, they are not always clustered and only Bacillus subtilis contains all three genes in the same order as

Streptomyces sp. D7*. Other microorganisms contain vdcB and vdcC homologues, but at different chromosomal locations. Interestingly, Sphingomonas aromaticivorans strain

F199 possesses homologues to vdcB and vdcC and vdcD on its 184 kb catabolic plasmid pNLl (Romine et al, 1999, GenBank accession AF079317). Plasmid pNLl contains a variety of genes encoding enzymes for the degradation of a number of toxic organic chemicals, and among these genes lie (in order; identifications by similarity only): orfl244 (vdcB, phenylacrylate decarboxylase), pchFa (p-cresol methylhydroxylase), vdh

(vanillin oxidoreductase), orfl272 (vdcC), orfl280 (vdcD).

* Interestingly, the Bacillus subtilis strain did not metabolize vanillic acid. 54

SD7 2 YDDLRYFLDTLEKEGQLL 19 I III ll+ I + I l+ CHB 6 YRDLREFLEVLXQXGXLI 2 3

Figure 14: Amino-terminal sequence alignment of Streptomyces sp. D7 protein 3717 (SD7) and Clostridium hydroxybenzoicum /7-hydroxybenzoate carboxy-lyase (CHB) (He and Wiegel, 1995). (|) = identity; (+) = similarity. Note that three amino acid residues of the C hydroxybenzoicum enzyme were not identified (marked as 'X'). 55

Figure 15: Lambda DASH II phage clone carrying the putative vanillate decarboxylase gene. Phage plaque lift was hybridized to 32P-labeled oligonucleotide probe 3717C and exposed to X-ray film for several hours with an intensifying screen. An arrow indicates the position of the hybridizing phage clone. 56

Figure 16: DNA purified from phage clone 3717C(+) and digested with BamHI (Lane 1). Molecular size markers (Lane M) are denoted in kilobases. Streptomyces sp. D7 genomic DNA fragments were separately purified and ligated into pUC19 to form the pKCE series of plasmids. The gel consists of 0.7% agarose in TAE buffer, stained with ethidium bromide. 57

M1 M2 A B C D E

Figure 17: pKCEl (Lane A), pKCE3 (Lane B), pKCE4 (Lane C), pKCE5 (Lane D) and pKCE6 (Lane E) purified plasmid DNA, digested with BamHl. An arrow denotes the position of linear pUC19. Ml = X-Hindlll molecular size standards, in kb; M2 = 1 kb molecular size standards, in kb. The gel consists of 0.7% agarose in TAE buffer, stained with ethidium bromide. 58

Lambda phage 3717C(+)

BamHI digestion, cloning of fragments into pUC19

pKCEl pKCE3 pKCE4 pKCE5 pKCE6 (7.1 kb) (5.7 kb) (4.9 kb) (4.6 kb) (3.5 kb)

Sa/I digestion, subcloning of 527 bp SD7 DNA fragment Sa/I cut pSUB1 DNA (527 bp)

pKCE2 (3.2 kb)

Figure 18: Plasmids carrying subfragments of the BamHI genomic DNA fragment from Streptomyces sp. D7, which was cloned in lambda DASH II phage clone 3717C(+). Plasmids are designated pKCEl through pKCE6, and are described in the text. 59

1 ccgcgtccag atcgctccgt agcgtgaaag cacaggtcag cgacaggttt gaggtggtcc vdcB-» 61 tf|||cggttg gtcgtgggaa tgaccggggc gacgggtgcc ccgttcggtg tccgtcttct 121 ggagaatctg cgccagttgc cgggcgtgga gacacatctc gtcctgtccc gctgggcgcg 181 caccaccatc gagatggaga cgggcctgtc cgtggccgag gtgtccgccc tagcggacgt 241 cacgcaccac cccgaggacc agggcgccac catctcctcc ggttcgttcc gcaccgacgg 301 catggtgatc gtgccgtgct ccatgaagac cctcgccggg atcaggaccg gatacgccga 3 61 agggctcgtc gcccgggccg ccgacgtggt tctcaaggag cgacgcaggc tcgtcctcgt 421 cccgcgcgag acaccgctga gcgagatcca cctccagaac atgctcgaac tcgcccgcat 481 gggcgtgcaa ctggtgccgc ccatgcccgc cttctacaac aacccgcaga ccgtcgacga 541 catcgtcgat cacgtggtgg cccgcatcct cgaccagttc gacctgcccg cgcccgccgc 601 ccggcgctgg gccgggatgc gcgccgcccg cgccgccgcc cgatccttcg gcgacgccgc vdcC-> 661 c||||§agtccc ctgccccgta aaggaattca cc|S||gccta tgacgacttg cgcagcttcc 721 tcgacacctt ggagaaggag gggcagctgc tgcgcatcac cgacgaggtg ctgcccgagc 781 cggatctcgc ggcggccgcc aacgcgaccg gccgcatcgg cgagaacgcc cccgccctcc 841 acttcgacaa cgtcaagggc ttcaccgacg cccgcatcgc gatgaacgtg cacggctcct 901 gggccaacca cgcgctcgcg ctcgggttgc cgaagaacac gccggtcaag gagcaggtgg 961 aggagttcgc gcggcgctgg gacgccttcc ctgtcgcccc cgagcgccgc gaggaagcac 1021 cctggcgtga gaacacccag gagggcgagg acgtcgacct gttctcggtg cttcccctct 1081 tccgcctcaa cgacggagac ggtggcttct a'tctcgacaa ggccgccgtc gtctcccgcg 1141 acccggagga ccgggacgac ttcggcaagc agaacgtcgg cacctaccgc atccaggtca 1201 tcggcaccaa ccggctcgcc ttccaccctg ccatgcacga cgtggcccag catctgcgca 1261 aggccgagga gaagggcgag gacctgccca tcgccatcac cctcggcaac gaccccgtga 1321 tggcgatcgt ggccgggatg ccgatggcgt acgaccagag cgagtacgag atggcgggag 1381 ccctgcgcgg cgcgcccgcg cccatcgcca ccgcgccgct caccggcttc gacgtgccct 1441 gggggagcga ggtcgtcata gagggcgtca tcgagtcccg caagcgccga atagaggggc 1501 ccttcggcga gttcaccggt cattactcgg gggggcgccg catgcccgtc atccgcgtgg 1561 aacgcgtctc gtaccggcac gaaccggtct tcgagtcgct ctacctcggc atgccgtgga 1621 acgagtgcga ctacctcgtc ggacccaaca cgtgcgtgcc gctgctcaag cagctgcgcg 1681 ccgagttccc cgaggtgcag gccgtcaacg ccatgtacac gcacggcctg atggtgatca 1741 tctccacggc caagcggtac ggcggcttcg ccaaggccgt cggcatgcgc gccatgacga 1801 cgccgcacgg gctcggctac gtggcccagg tgatcctcgt cgacgaggac gtcgacccgt 18 61 tcaacctgcc gcaggtcatg tgggcgatgt ccgccaaggt caacccgaag gacgacgtcg 1921 tcgtcatccc caacctgtcg gtcctggaac tcgcgcccgc cgcgcagccc gccggcatca 1981 gcagcaagat gatcatcgat gcgacgacgc cggtcgcccc ggacgtccgc ggcaacttct 2041 ccactccggc caaggacctg cccgagaccg cagagtgggc cgcccgcctg cagcgcctga vdcD -> 2101 tcgcggcccg cgtc|||jlcac ccaccgatcc tcactgaaag gacaactccc gjj§aaccacc 2161 tgcccgttga atgcccccgc tgcgcctttg aggacatctc cctgcttgcc acgtcccccg 2221 tcccaggcgt gtgggacgtg gtccagtgcg gccgctgtct ctacacctgg cgcacgatcg 2281 aacccgcacg ccgcacccgg cgtgacgcct acccggacag cttcaagctg acggcggagg 2341 acatcgagaa cgccatcgag gtgcccgcgg tgccgccact gctcaagH!§ cgttcctggc 2401 acgcccggag agaacgtccg ccagaacctg aacgggcccg gcgtcgagct cccacatggg 24 61 agtgagggcg ggtggtgccc aactttccct catatacacg tagttgctac tcgctcgcac 2521 tccgtctcgc gcagtgaacg tagaaaatat tcggtagtag gtagtgcaac ggaactcatt 2581 gacgcgctac ctttcgcctg ccccgtcacg gcaggcacac caacggtgtt gggccccatc 2 641 cgctgaccgg ggccgaaggc gcacagttcc tgtccaaccc agagactgaa tctggatgca 2701 tgacaccgct ccgttcctcg ctgcacgaat gcgcgggcaa ccaccgttct ggccgtactg 2761 gccgcgctgg gcacctctgt cgtggctgcc gccgcgtcgc ccgccgcggc caaggacgtc 2821 tcgtatcgcg gttatcacgt acgcgtgccc gcaagttggc cagtcgtgga cctgaccgcg 2881 aaccccggca cctgcgtgcg cctcgaccgg cacgcggtct atgtcggtca tccctccgac 2941 gccggtcagg cctcgtgccc g

Figure 19: Nucleotide sequence of the vdc gene cluster, featuring vdcB, vdcC and vdcD. Putative Shine-Delgarno sites have been underlined, and start (ATG, GTG) and stop (TGA) codons are JCAPTAUZLD IN 130LI) A\'p"sTlAI)l;.p. for each open reading frame. 60

RBS RBS RBS 1 I 1 — vdcB vdcC vdcD //

BamHI BamHI 0.6 kb 1.4 kb 0.2 kb

Figure 20: Schematic diagram of the 4.4 kb BamHI genomic DNA fragment from Streptomyces sp. D7, containing the VDC gene cluster. Putative ribosome binding site (RBS) locations have been indicated with arrows. 61

Figure 21 (following pages): Amino acid sequence alignments between the translation products of vdcB (a) and vdcC (b) and their respective putative homologues from various microorganisms. Dendrograms depicting these alignments are shown for the products of the vdcB gene (c) and the vdcC gene (d). These dendrograms and sequence alignments were produced using the Pileup program, which is part of the Wisconsin Package Version 10 bioinformatics software package (GCG). eco: Escherichia coli; sph: Sphingomonas aromaticivorans; bsu: Bacillus subtilis; str: Streptomyces sp. D7; scv: Saccharomyces cerevesiae; arc: Archaeoglobus fulgidus; meb: Methanobacterium thermoautotrophicum; mec: Methanococcus jannaschii; pyr: Pyrococcus horikoshii; aqx: Aquifex aeolicus; bfi: Bacillus firmis; chl: Chlamydia trachomatis; hpy: Helicobacter pylori; mbr: Methanobrevibacter smithii; rho: Rhodospirillum rubrum; syn: Synechocystis PCC6803; PAD: homologue of phenylacrylate decarboxylase; C: homologue of vdcC The region of the vdcB homologues that is a putative membrane association domain is jholrit-ri anil 62

(a)

ecoPAD : "LIVGISq sphPAD • MWGITG bsuPAD -MKAEFKRKGGG • LWGMTG strPAD -L.WGMTG scvPAD MLLFPRRTNIAFFKTTGIFANFPLLGRTITTSPSFLTHKLSKEVTRASTSPPR ;WAITG arcPAD n-KWAIiTG mebPAD -~~MKV :'"|1TIAMTG me c PAD -, lnvcITG pyrPAD '-IIVAITG aqxPAD ; bfiPAD -MSERIEKIMTVGITGj chlPAD :.ywGISG hpyPAD MKJVLGISG 10 20 30 40 50 60 ecoPAD ASGAIYGVR .LQVLRDVTDIETHLVMSQAARQTLSLETDFSLREVQALADV TH sphPAD ATGSVYGLR7.LELLRETGGWETHLVMSPAALLNIREELPEGKARLEALADV VH bsuPAD ATGAIFGVR .LQWLK.AAGVETHLVVSPWANVTIKHETGYTLQEVEQLATY TY StrPAD ATGAPFGVSI'JLENLRQLPGVETHLVLSRWARTTIELETGLSVAEVSALADV TH s cvPAD ATGVALGIR.iLQVLKEL . SVETHLVISKWGAATMKYETDWEPHDVAALATK TY arcPAD ASGQILGIBL.JIEKLTEL . GAEVYAVASRAAKITLKAETDYDEGYVREIATK YY mebPAD ASGVIYGEEiiLKALRGA.GVRIGLMITDTAREIIRYELGIEPGALEELAD E. .CF mecPAD ASGVIYAKR iLEVLKDR. .AEVNLIISNSAKKIIKEELDIDWKEIKKLAT D. .YY pyrPAD ASGSIYGIK .YEILKKL. GHDVILLASKTGIKVAKYETGME IT P. .DF aqxPAD ASGVIYGIK.iLQVLEEL . DFSVDLVISRNAKWLKEEHSLTFEEVLKGLK NVRIH bfiPAD ASGGMYGVR-.TQELLRQ. EYKVHLVLTEAAWQVFKEELLLDTTDRQKVIHELFGDLPGELHTH chlPAD ASGIVLAVfl .VSELARL . GHHIDVIISPSAQKTLYYEL DTKSFLSTIPQNFHNQIVLH hpyPAD ASGIPLAIggJFLEKLPK . . EIEVFWASKNAHWALEESNINLKNAMK DLRPSGTFF 70 80 90 100 110 120 ecoPAD DARDIAASISSGSFQT..LGMVILPCSIKTLSGIVHSYTDGLLTRAADWLKERRPLVLCVRE sphPAD NVRNVGASIASGSFVC..EGMAIAPCSMRTLGAVAHALSDNLITRAADVMLKERRRLVMITRE bsuPAD SHKDQAAAISSGSFDT..DGMIVAPCSMKSLASIRTGMADNLLTRAADVMLKERKKLVLLTRE StrPAD HPEDQGATISSGSFRT..DGMVIVPCSMKTLAGIRTGYAEGLVARAADWLKERRRLVLVPRE scvPAD SVRDVSACISSGSFQH..DGMIWPCSMKSLAAIRIGFTEDLITRAADVSIKENRKLLLVTRE arcPAD DEDEIAAPFASGSFRH..DGMAVVPCSIKTASSIAYGIADNLIARAADVTLKEKRRLVLAIRE mebPAD DASDFTTSINSGSS..PFRAMVIAPCTMKTLSAIANGYAENSLTRAADVCLKERRDLVLVPRE mecPAD ENDDFFSPLASGSN..KFDAVWVPCSMKTLSAIANGYSANLIVRVCDIALKERRKLIIMPRE pyrPAD DEDDLFAPIASGSY..PFDAMVIAPCSMKTLGAIANGFSYNLITRAADVTLKERRKLILLIRE aqxPAD EENDFTSPLASGSRLVHYRGVYWPCSTNTLSCIANGINKNLIHRVGEVALKERVPLVLLVRE bfiPAD DLHDYAAPIASGSYRSA..GMVIIPCSMGTLSGMAHGASGNLLERTADVMLKEKRKLVIVPRE ChlPAD HISSIESSVSSGS..NTIDATIIVPCSVATVAAISCGLADNLLRRVADVALKEKRPLILVPRE hpyPAD NEQDIHASIASGSY..GIHKMAIIPASMDMVAKIAHGFGGDLISRSASVMLKEKRPLLIAPRE 130 140 150 160 170 180 ecoPAD TPLHLGHLRLMTQAAEIGAVIMPPVPAFYHRPQSLDDVINQTVNRVLDQFAITLPEDLFARWQ • sphPAD APLNLAHLRNMTACTEMGAVIFPPVPAFYARPTSLADWDHTCMRVLDLFGLHAKSE..KRWQ bsuPAD TPLNQIHLENMLALTKMGTIILPPMPAFYNRPRSLEEMVDHIVFRTLDQFGIRLPE..AKRWN strPAD TPLSEIHLQNMLELARMGVQLVPPMPAFYNNPQTVDDIVDHWARILDQFDLPAPA..ARRWA scvPAD TPLSSIHLENMLSLCRAGVIIFPPVPAFYTRPKSLHDLLEQSVGRILDCFGIHADT..FPRWE arcPAD APLHSGHLKTLARLAEMGAVIFPPVLSFYTRPKSVDDLIEHTVSRIAEQLGVEVDYRRWG mebPAD TPLRSVHLENMLRVSREGGIILPAMPGFYHKPASIEDMADFIAGKVLDVLGI..ENDLFRRWT mecPAD MPFNSIHLENMLKLSNLGAIVMPPIPAFYNKPKNVNDIINFVVGRVLDILGI..DNSLFKRWG pyrPAD TPLNLVHVQNMLKIIQAGGIIMPASPAFYTKPKTIDDMVNFIIGKILDLLGI..THNLYRRWG aqxPAD APYNEIHLENMLKITRMGGVWPASPAFYHKPQSIDDMINFWGKLLDVLRI. . EHNLYKRWR bfiPAD TPLHDIHLENMLKLSKMGATILPAMPGYYHLPKTIDDLINFLVGKALDSLGV..EHTLFTRWG ChlPAD APLSAIHLENLLKLAQNGAVILPPMPIWYFKPQTAEDIANDIVGKILAILQL..DSPLIKRWE hpyPAD MPLSAIMLENLLKLSHSNAIIAPPMMTYYTQSKTLEAMQDFLVGKWFDSLGI..ENDLYPRWG 190 200 210 220 230 240 250 ecoPAD GA sphPAD GLSKEAASLVPGAGQMEGN bsuPAD GIEKQKGGA- StrPAD GMRAARAAARS FGDAA SCVPAD GIKSK ~ mebPAD GKDI mecPAD TV pyrPAD MREDD ~ aqxPAD G bfiPAD E ~ chlPAD NPR hpyPAD MN 260 270 280 290 300 310 64

(b)

mebC ~ -~~ MRNFLDKIGEEALV mbj-C . _—„ MDIKDENIIE mecC ~~~ —~ MREIINKLNP. II pyrC MVMKMLREIVES FEDL W rhoC ~ ~ ~~~~MERDFSGSPRVIADLGRIIDRLEALGRLVR bsuC MAYQDFREFLAALEKEGQLLTVNEEVKPEPDLGASARAASNLGDKSPALLFNNIY strC MAYDDLRS FLDTLEKEGQLLRITDEVLPEPDLAAAANATGRIGENAPALHFDNVK sphC- MTMNDLPNRARSISSLRDFLELLEDAGQAITWSDAVMPEPGVRNIAVAASRDANGAPAIVFDNIT aqxC MGYKYRDLHDFIKDLEKEGELVRIKEPLSPILEITEVTDRVCKMPGGGKALLFENP.

arcC MAYEDLREFIGRLEDKGELARVKHEVSPILEMSEVADRTVK. . AGGKALLFERP . ecoC MDAMKYNDLRDFLTLLEQQGELKRITLPVDPHLEITEIADRTLR..AGGPALLFENP. synC MARDLRGFIQLLETRGQLRRITAEVDPDLEVAEISNRMLQ. .AGGPGLLFENV. hpyc MRDFLKLLKKHDELKIIDTPLEVDLEIAHLAYIEAKKPNGGKALLFTQPI chlC MFSLRSLVDYLRSQHELIDIHVPVDPHLEIAEIHRRVVERE. . GPALLF. . . . 10 20 30 40 50 60

mebC VEDEVSTSFEAASILREHPRDL..VILKNLKESDIPVISGLCNTREKIALSLNCRVHEITHRIVE mbrC ITTELSSEFEVAKELRKYPKDT..VIIKNVKGYDLPIISGICNTREKIAKSINCEVSEITQKIIE mecC IDKADKK. FGVSRILKKYDGKP. . VYIKDVNGFE . . WGNL . CSRETLSKIFNVKKEDFIFFMLD pyrC IDKPVKKELELTKFLLKYKDKP..VLFKDVEGWE..VAGNLWSSRERIAKFLNTDNKGLLELLYE rhoC VRSEVDPRHDLAGIAARFEGGPQAVLFEKVAGHAYPVFVGLYWSRELLGALFDQPETALPQHVAA bsuC GYHNAR IAMNVIGSWPNHAMMLGMPKDTPVKEQFFEFAKRY StrC GFTDAR IAMNVHGSWANHALALGLPKNTPVKEQVEEFARRW sphC GYPGKR LAVGVHGSWDNIALLLGRPKGTTIRELFFEIAGRWGD aqxC KGYRIPVLTNLYGSEKRIKKALGYEN . . . LEDIGWKLYRILKPEVPKTFLEKIKKLPE arcC KGYDIPVFMNAFGTERRMKLALEVER...LEEIGERLLSALEFR.PSSFMDALKGVGM ecoC KGYSMPVLCNLFGTPKRVAMGMGQEDVSALREVGKLLAFLKEPEPPKGFRDLFDKLPQ synC KGSPFPVAVNLMGTVERICWAMNMDHPLELEDLGKKLALLQQPKPPKKISQAIDFGKV hpyC RKEHDQIKTFGMPVLMNAFGSFKRLDLLL....KTPIESLQQRMQAFLHFNAPKNFTEGLKVLKD ChlC ....HQVKGSPFPVLTNLFGTRRRVDLLFPDLSSDLFEQIIHLLSS PPSFSSLWKHRSL 70 80 90 100 110 120 130

mebC AMENP....T..PISSVGGLDGYRSGRADLSELPILRHYRRDGGPYITAGVIFARDPDT....GV t mbrC ASDNP. ... I. .KVDKFTDFSDYNTTEANLDKIPILTHYKRDGGKYITAGWFARDPET. . . .GI ' mecC - AMEKEKEGKL. . KINNKLKEKYIVEIPENIKNWPIPIYYEKDAGAYITSGWWYDKDY . . . . GY pyrC AMEKPKPFSV. .VEKAEFLKN...REKVNLLELPIPKYYPKDGGPYLTSAMVIA..KKE FV rhoC SIKSWQSAPVDPLWADGPVLEVTEAEVDLSTLPIPIHALEDGGPYFDAAWIAKDPET . . . . GV bsuC ..DQFPMPVKREETAPFHEN.EITEDINLFDILPLFRINQGDGGYYLDKACVISRDLEDPDNFGK StrC ..DAFPVAPERREEAPWRENTQEGEDVDLFSVLPLFRLNDGDGGFYLDKAAWSRDPEDRDDFGK sphC ..QEAQISFVPEAQAPVHE.CRIEQDINLYDVLPVYRINEYDGGFYIGKASVASRDPLDPDNFGK aqxC LKKLNDAIPKWKRGKVQEEVIMGD.INLED.LPILKCWPKDGGRYITFGQVITKDPES....GI arcC LKDFMSFIPK. . KTGKAPCKEWAE . . SLDK. FPILKCWPKDAGRFITFPWITKDPET . . . . GE ecoC FKQVLNMPTKRLRGAPCQQKIVSGDDVDLNR.IPIMTCWPEDAAPLITWGLTVTRGPHK....ER synC LFDVLKAKPGRNFFPPCQEWIDGENLDLNQ.IPLIRPYPGDAGKIITLGLVITKDCET....GT hpyC LWDLRHIFPKKTTRPK.DLIIKQDKEVNLLD.LPVLKTWEKDGGAFITMGQVYTQSLDH....QK ChlC FKRGISALGMRKRHLR.PSPFLYQDAPNLSQ.LPMLTSWPEDGGPFLTLPLVYTQSPEN....GV 140 150 160 170 180 190

mebC RNASIHRMMVIGDDRLAVRI.VPRHLYTYLQK..AEERGEDLEIAIAIGMDPATLLATTT...SI mbrC QNASIHRMLVLDDKRLVIRI.VPRNLYTYFQK..AQKLGKDLEIAIAIGMDPAILLASTT...SI mecC .NLSIHRILVKDDYLVIRMV.EQRHLHFLYNK..ALKEKGYLDVAIVIGVHPAVLLAGST...SA pyrC .NVSFHRMMVLDEERAVIRL.VPRHLYSMWKD..SVEHGEELEVRIVLGNPVHLLLAGAT...SV rhoC RNASIQRFQVIGKDRLVINIDAGRHLGLYLDK..AAARGEPLAFTLNVGVGPGVHFAAAAPAEAA bsuC QNVGIYRMQVKGKDRLGIQPVPQHDIAIHLRQ..AEERGINLPVTIALGCEPVITTAASTP...L StrC QNVGTYRIQVIGTNRLAFHPA.MHDVAQHLRK..AEEKGEDLPIAITLGNDPVMAIVAGMP...M sphC QNVGIYRLQIQGPDTFTLMTIPSHDMGRQIMA..AEREGVPLKIAVMLGNHPGLAAFAATP...1 aqxC RNVGLYRLQVLDKDKLAVHWQIHKDGNHHYWK..AKRLGKKLEVAIAIGGEPPLPYVASAP...L arcC MNAGMYRMQVFDGKTTGMHWQIHKHGAEHFRK.MAEKGGGKIEVAVAIGVDPATLYAATAP...L ecoC QNLGIYRQQLIGKNKLIMRWLSHRGGALDYQEWCAAHPGERFPVSVALGADPATILGAVTP...V 200 210 220 230 240 250 260 65

synC PNVGVYRLQLQSKTTMTVHWLSVRGGARHLRK..AAEQGKKLEVAIALGVDPLIIMAAATP...1 hpyC KNLGMYRLQVYDKNHLGLHWQIHKDSQLFFHEYAKAKV..KMPVSIAIGGDLLYTWCATAP...L chlC PNLGMYRMQRFDKETLGLHFQIQKGGGAHFFE..AEQKKQNLPVTVFLSGNPFLILSAIAP...L 200 210 220 230 240 250 260 mebC PIDADEMEVANTFH....EGELELVRCEGVDMEVPPAEIILEGRILCGVRE.REGPFVDLTDTYD mbrC PIDYNEMDVANAFK....NGELTLIKCG..DLEVPQADIILEGKISVSETS.AEGPFVDLTDTYD mecC DITFDELKFAAAL....LGGEIGVFELDNGLL.VPEAEFIIEGKIL.PEVD.DEGPFVDITGTYD pyrC AYGVSELEIASAISLKAFGRPLEVINLDGIPT.PVDSEFVFKAKIT.DEVA.DEGPFVDITGTYD rhoC PVETDELGIASAFHGAPLELVAGTV...GPVEMVAHAMWALECEIRPGEVH.AEGPFAEVTGYYA bsuC LYDQSEYEMAGAIQGEPYR.IVKSKLSD..LDVPWGAEWLEGEIIAGEREY.EGPFGEFTGHYS strC AYDQSEYEMAGALRGAPAP.IATAPLTG..FDVPWGSEWIEGVIESRKRRI.EGPFGEFTGHYS sphC GYEESEYSYASAMMGAPIR.LTKSG.NG..IDILADSEIVIEAELQPGGREL.EGPFGEFPGSYS aqxC PPEVDEYLFAGIIMERPVE.LVKGLTVD..LEYPANAEIAIEGYVDPEEPLVDEGPFGDHTGFYT arcC PSGISEFMFAGFIRKERLK.VTECETVD..LLVPANAEIILEGYVRVDEMRV.EGPFGDHTGYYT ecoC PDTLSEYAFAGLLRGTKTE.WKCISND..LEVPASAEIVLEGYIEQGET.APEGPYGDHTGYYN synC PVDLSEWLFAGLYGGSGVA.LAKCKTVD..LEVPADSEFVLEGTITPGEM.LPDGPFGDHMGYYG hpyC PYGIYELMLYGFMRGKKAR.VMPCLSNS..LSVPSDCDIVIEGFVDCEKLEL.EGPFGDHTGYYT ChlC PENVPELLFCSFLQNKKLSFVEKHPQSG..HPLLCDSEFILTGEAVAGERR.PEGPFGDHFGYYS 270 280 290 300 310 320 mebC VVRDEPVISLERMHIRKD.AMYHAILPAGF.EHRLLQGLPQEPRIYRAVKNTVPTVRNVVLTEGG mbrC IIRDQPIINLSKMHIKKDNPHYHGILSAGF.EHKLLQGLPQEPRIFKSVKNAVPTVENWLTEGG mecC IVRKQPIIKIEKLY.RKEKPIFHALLPGGI.EHKTLMGMPQEPRILKGVRNTVPTVKNIVLTEGG pyrC IVRKQPIVIFEEMY. HVDDPIFHALLPGGY. EHYMLMGLPKEPQIYASVKKWPKVHGVRLTEGG rhoC RVEPRPLVRVKRIH.RRRAPIFHTLL.SGA.EVFNSVGLLGEANVLALLRVQVPGVEDVYFSHGG bsuC GGRSMPIIKIKRVYHRNNPIFEHLYLGMPWTECDYMIGINTCVPLYQQLKEAYPNEIVA.VNAMY StrC GGRRMPVIRVERVSYRHEPVFESLYLGMPWNECDYLVGPNTCVPLLKQLRAEFP.EVQA.VNAMY sphC GVRKAPIFKVTAVSHRRDPIFENIYIGRGWTEHDTLIGLHTSAPIYAQLRQSFP.EVTA.VNALY aqxC PVDKYPQMHVTAIVMRKDPIYLTTIVGRPPQE.DKYLGWATERIFLPLIKFNLPEWDYHLPAEG arcC PPEPYPVFHITHITHRENPIYHATWGKPPME.DAWLGKATERIFLPILRMMHPEIVDINLPVEG ecoC EVDSFPVFTVTHITQREDAIYHSTYTGRPPDE.PAVLGVALNEVFVPILQKQFPE1VDFYLPPEG synC GVEDSPLVRFQCLTHRKNPVYLTTFSGRPPKE.EAMMAIALNRIYTPILRQQVSEITDFFLPMEA hpyC PIEPYPVLEVKTISYKKDSIYLATWGKPPLE.DKYMGYLTERLFLPLLQMNAPNLIEYYMPENG ChlC LTHDFPIFKCNCLYHKKDAIYPATWGKPFQE.DFFLGNKLQELLSPLFPLIMPGVQDLKSYGEA 330 340 350 360 370 380 390 mebC CCWLHAAVSIKKQTEGDGKNVIMAALAAHPSL. . KHVVWDEDIDVLDPEEIEYAIATRVKGDD mbrC CCWLHAAISINKQTEGDGKNAIMAALSAHPSL. , . KHAWVDTDVDVFDPQDIEYAIATRVKGDR mecC CCWLHAWQIEKRTEGDGKNAILAAFASHPSL. . . KHVIWDDDINIFDINDVEYAIATRVQGDK pyrC CMWLHAWSITKQHEGDGKNAILAAFAGHPSL. ,. KRWWDEDVNIYDDREVEWAIATRFQPDR rhoC CGFYHCWKIAQKRAGWAKQAILATFAAFPPL. , . KMVTWDEDVDIRNGRDVEWAMTTRLDAKT bsuC THGLIAIVSTKTRYGGFAKAVGMRALTTPHGLGYCKMVIWDEDVDPFNLPQVMWALSTKMHPKH StrC THGLMVIISTAKRYGGFAKAVGMRAMTTPHGLGYVAQVILVDEDVDPFNLPQVMWAMSAKVNPKD sphC QHGLTGIISVKNRMAGFAKTVALRALSTPHGVMYLKNLIMVDADVDPFDLNQVMWALSTRTR.AD aqxC. CFHNFCFVSIKKRYPGHAFKVAYALLG.LGLMSLEKHIWFDDWINVQDIGEVLWAWGNNVDPQR arcC AFHNLAIVSIKKRYPGQAKKVMYAIWG. TGMLSLTKIVWVDDDVNVHDMREWWAVTSRFDPAR ecoC CSYRLAWTIKKQYAGHAKRVMMGVWSFLRQFMYTKFVIVCDDDVNARDWNDVIWAITTRMDPAR s ynC LSYKAAIISIDKAYPGQAKRAALAFWSALPQFTYTKFVIVVDKSINIRDPRQWWAISSKVDPVR hpyC VFHNLILAKIHTRYNAHAKQVMHAFWGVGQMSFVKHAIFVNEDAPNLRDTNAIIEYILENF..SK ChlC GFHALAAAIVKERYWKEALRSALRILGE.GQLSLTKFLWITDQSVDLENFPSLLECVLERMNFDR 400 410 420 430 440 450 mebC DILIVPGARGSSLDPAA.LPDGTTTKVGVDATAPL.ASAEKFQRVSRSE mbrC DLMIVPNVRGSSLDPVA.ESDGTTTKIGLDATKSL.KTLDKFERVSFGE mecC DIVIISGAKGSSLDPSSDLKNKLTAKVGVDATMSLIKGREHFERAKIPDK pyrC DLVIIPNARGS SLDPSG..KDGLTAKWGIDATKPLDKKKE.FEKASLDF rhoC GILVIENAFGHGLNPT..FPNYLGTKVGFDCTRPFPHT.PAFDRAKTKAMTLDGLDIVGAKR bsuC DAVIIPDLSVLPLDPGSNPSGITHKMI.LDATTPVAPETRG.HYSQPLDSPLTTKEWEQKLMDLM 460 470 480 490 500 510 520 66

StrC DVWIPNLSVLELAPAAQPAGISSKMI.IDATTPVAPDVRG.NFSTPAKDLPETAEWAARLQRLI sphC DIIVLPNMPAVPIDPSAWPGKGHRLI.IDATSYLPPDPVG.EAHLVTPPTGDEIDALSKRIREM aqxC DVLIL.KGPIDVLDHATNEVGFGGK.MIIDATTKWKEEGYTREWPEVIEMSPEVKKRIDEIWDRL arcC DVVILPPSPTDSLDHSAYIPNLAGK.LGIDATKKWRDEGYEREWPDWEMDAETKRKVDAIWNEI ecoC DTVLVENTPIDYLDFASPVSGLGSK.MGLDATNKWPGE.TQREWGRPIKKDPDVVAHIDAIWDEL synC DVFILPETPFDSLDFASEKIGLGGR.MGIDATTKIPPE.TDHEWGEVLESDPAMAEQVSQRWAEY hpyC ENALISQGVCDALDHASPEYAMGGK.LGIDATSK SNTPYPTLLNDSALLALLQDKMQNIV chlC DLLILSETANDTLDYTGSGFNKGSKGIFLGVGAPIRSLPRRYRGPSLPGISQIGVFCRGCLVLET 460 470 480 490 500 510 520 bsuC NK : strC AARV sphC QLGALS aqxC GIE arcC RNMVL ecoC AIFNNGKSA synC GLGDINLTEVNPNLFGYDV hpyC LLKQYYPHTRNPICVISVEKKDKSVIELAKNLLGFEEHLRIVIFVEHASNDLNNPYMLLWRIVNN ChlC SLQQLDIPALLKEPHLADWPLVILVEDLSSALSSTKEFIWRTFTRSSPATDLHIPVSQITNHKVS 530 540 550 560 570 580 hpyC IDARRDILTSKHCFFIDATNKGVMDKHFREWPTETNCSMEVIENLKKKGLLKDFETLNQKFHLTH ChlC YTPPMILNALMKPPYPKEVEADEATQNLVSSRWHSYFP— 590 600 610 620 630 640 650 hpyC SFSTHKEDL ~ 660 670 680 690 700 710 67

Escherichia coli

Sphingomonas aromaticivorans

Bacillus subtilis

Streptomyces sp. D7

Saccharomyces cerevisiae

Archaeoglobus fulgidus

Methanobacterium thermoautotroph

Pyrococcus horikoshii

Methanococcus jannaschii

— Aquifex aeolicus

Bacillus firmus

Helicobacter pylori

Chlamydia trachomatis Methanobacterium (d) thermoautotrophicum

Methanobrevibacter smithii

Methanococcus jannaschii

Pyrococcus horikoshii

Rhode-spirillum rubrum

I— Bacillus subtilis

I Streptomyces sp. D7

' Sphingomonas aromaticivorans

' Synechocystis PCC6803

I Escherichia coli

' Chlamydia trachomatis

r—— Aquifex aeolicus

I Helicobacter pylori

_ Archaeoglobus fulgidus 69

3.2 Detection of a putative transcriptional regulatory gene

Preliminary DNA sequencing of the region upstream of the VDC gene cluster indicates the presence of a putative divergently transcribed regulatory gene, the position of which is illustrated in Figure 22. BLAST-X analysis of the translation product of the region matched strongly to a putative positive transcriptional activator from the Streptomyces coelicolor A3(2) genome (Figure 23). 70

BamHI BamHI

4.4 kb BamHI fragment in pSUBl

Figure 22: Location of the putative divergent regulatory gene in relation to the VDC gene cluster. 71

embICAA19943| (AL031107) putative transcriptional regulator [Streptomyces coelicolor] Length = 301

Score = 119 bits (295), Expect = 5e-26 Identities = 86/256 (33%), Positives = 111/256 (42%), Gaps = 4/256 (1%) Frame = -1

Query: 1180 FHQLIGVRGRPLRMWVDFVEHELRPGSWLWIRPGQVQRFGPDLAAADGVIVLFQPGFLPP 1001 FH ++ G P+R +DF E+E G LWIRPGQV RF P+ G ++ QPGFLP Sbjct: 54 FHIVMLFTGGPVRHMIDFAEYEASAGDLLWIRPGQVHRFAPE-GEYRGTVLTMQPGFLPR 112

Query: 1000 TTVSLAHMDPPYE-QRPSVLEGSDAE—GVRRALDHLVHEYGAMASLPLQAHTE-XXXXX 833 TV + Y P +L +A G+ AL+ L EY. +LPL HT Sbjct: 113 ATVEATGL YRYDLPPLLHPDEARLAGLTAALEQLRREYEDATTLPLSLHTAVLRHTL 169

Query: 832 XXXXXXXXXXXRGPAPRPTAAFGNVPSLSHXXXXXXXXXXGASTNTRGRSATAPHLTRRV 653 G A PS G +TN V Sbjct: 170 SAFLLRLAHLAAGSARAARQGRAEAPGDSTFVLFRDAVERGFATN HSV 217

Query: 652 DDYAAALGYSSXXXXXXXXXXXXXXANQYVDDRVLLEAKRLLQHSGLTAREVTVRLGFTD 473 YA ALGYS ++D RV+LEAKRLL H+ + V +GF D Sbjct: 218 SAYADALGYSRRTLVRAVRAATGETPKGFIDKRVVLEAKRLLAHTEMPIGRVGAAVGFPD 277

Query: 472 ASDFTKFFRLRTGMTPGAFR 413 A++F+KFF+ T TP AFR Sbjct: 278 AANFSKFFQQHTDQTPAAFR 297

Figure 23: BLAST-X result indicating that the nucleotide region immediately upstream of vdcB encodes a polypeptide similar to a putative transcriptional regulator from S. coelicolor A3(2). Query = Streptomyces sp. D7 translated nucleotide sequence; Subject = S. coelicolor A3(2) genomic DNA translation product. 72

3.3 Messenger RNA analyses

Messenger RNA was isolated from Streptomyces sp. D7 under both uninduced and vanillic acid-induced conditions, blotted to a positively charged nylon membrane, then probed with 32P-labeled vdcC PCR amplification product. The resulting autoradiogram revealed that a transcript of approximately 2.3 kb was synthesized in vanillic acid- induced cells (Figure 24). A transcript of this size corresponds to the expected length of all three VDC genes and their associated intergenic regions. This result indicates that the gene cluster may be transcribed from a single, inducible promoter. 73

Figure 24: Northern blot hybridization of PCR-amplified, radiolabeled vdcC DNA against mRNA isolated from uninduced Streptomyces sp. D7 (Lane 1) and Streptomyces sp. D7 induced with 3.6 mM vanillic acid (Lane 2). RNA size standards are indicated (Lane M). 74

3.4 Detection of VDC cluster genes in Streptomyces setonii 75Vi2

Southern hybridization of radiolabeled vdcB, vdcC, and vdcD PCR products to SaR- digested S. setonii 75Vi2 genomic DNA revealed that all three genes of the cluster appear to be present in the characterized vanillate decarboxylating strain (Figure 25). At the time of the writing of this report, a lambda DASH II phage clone of S. setonii 75Vi2 genomic DNA, which hybridizes strongly to radiolabeled vdcC probe, is being analyzed and sequenced (M.K. Pope, collaborative work) in order to further characterize the VDC cluster of 5*. setonii 75Vi2. Preliminary sequence information reveals that within the phage clone DNA insert, a cytochrome P-450 gene is located within several kilobases of the putative VDC gene cluster in the genome of S. setonii 75Vi2. In addition, the vdcB homologue in S. setonii 75Vi2 has been sequenced, with further sequencing underway to uncover the remaining genes. As mentioned earlier, cytochrome P-450 enzymes have been implicated in the demethylation of methoxyl groups on guaiacol by S. setonii 75Vi2

(Sutherland, 1986). No cytochrome P-450 gene was observed in the vicinity of the VDC gene cluster in Streptomyces sp. D7, which supports the observation that no transformation of guaiacol could be detected in the organism. These observations will be described in more detail in the Discussion section to follow. 75

Figure 25: Southern blot hybridization of Sa/I-digested chromosomal DNA from Streptomyces sp. D7 (Lane 1) and Streptomyces setonii 75Vi2 (Lane 2) with radiolabeled vdcB, vdcC and vdcD (as indicated) PCR-amplified DNA probes at 65 °C. Approximate molecular sizes (in kilobases) are shown on the left. 76

3.5 Gene knockout studies

Gene knockout experiments involve transforming cells with an inactivated form (allele) of the gene being studied, for homologous recombination with the wild type gene on the host's chromosome. Inactivation is usually achieved by replacing a section of the gene with an antibiotic resistance cassette. In streptomycete procedures, the thiostrepton resistance gene is commonly used. These experiments are commonly used for gene identification and/or functional analyses, and disruption and inactivation of the VDC gene cluster would allow determination of the functions of each of the three genes in vivo. In order to do this, Streptomyces sp. D7 had to be transformed with a disrupted allele of each gene, to obtain knockout mutants by double crossover recombination events with the chromosome. Protoplasts of Streptomyces sp. D7 were prepared, which regenerated efficiently on R6 agar plates designed for that purpose. However, in spite of many attempts, these protoplasts were not transformable with pIJ702 as a control plasmid using standard Streptomyces protoplast genetic manipulations. Likewise, attempts to introduce pPM801 into Streptomyces sp. D7 mycelia by interspecies Streptomyces-E. coli conjugation procedures failed to produce ex-conjugants expressing the kanamycin resistance marker gene of the vector. These observations suggest that Streptomyces sp.

D7 may possess a potent restriction-modification system, which degrades foreign DNA.

Streptomyces is particularly well known for being highly resistant to transformation procedures (Oh & Chater, 1997), which is the major reason why the vast majority of genetic research on these organisms has been conducted with either Streptomyces lividans or Streptomyces coelicolor, organisms for which transformation procedures have been well established. The resistance of Streptomyces sp. D7 to genetic transformation 77 made it necessary to devise alternative methods to characterize the functions of the VDC gene cluster. I decided to express each gene independently to examine their activity in

Escherichia coli and/or Streptomyces lividans 1326 as surrogate host systems, which is described in Section 3.6.

3.6 Gene expression

3.6.1 Expression with pET22b(+) in E. coli BL21(DE3)

Initially, the goal of the project was to clone individual VDC cluster genes behind the T7 polymerase promoter in the expression vector pET22b(+) (Novagen), to obtain high amounts of active enzymes in the expression strain Escherichia coli BL21(DE3). Studies of the purified, active protein would allow the deduction of the function of each enzyme in the catabolism of vanillic acid. However, although the vdcB and vdcC genes were successfully expressed in E. coli, the proteins were formed as inclusion bodies, no matter how the conditions for expression were varied to achieve proper protein folding. For example, protein expression was carried out at 37°C and 25°C, with both 1 mM and 0.1 mM IPTG used for induction of the T7 RNA polymerase system. Attempts were made to solubilize and refold the inclusion body material, but with no established biochemical assay and only a putative phenotype, it was decided that streptomycete vectors and host systems would be more appropriate for the expression of the VDC genes and the determination of their function(s). A typical example of SDS-PAGE-visualized protein over expressed using pET22b(+) in E. coli BL21(DE3) is shown in Figure 26, in which the vdcC gene product was produced as inclusion bodies at 37°C with 0.1 mM IPTG used 78 to induce protein synthesis. An example of inclusion body formation at lower induction temperatures is shown in Figure 27. 79

M 1 2

Figure 26: SDS-PAGE (12.5% acrylamide) of cell extracts showing the expression product of the vdcC gene, produced using pET22b(+) in E. coli BL21(DE3) at 37°C with 0.1 mM IPTG for induction. Lane 1: cell lysate from uninduced cells; Lane 2: cell lysate from IPTG-induced cells. The position of the vdcC expression product is denoted by an arrow. Molecular weight markers are shown on the left, in kilodaltons. 80

M 1 3 4

32.5

Figure 27: SDS-PAGE (12.5% acrylamide) of cell extracts showing the expression product of the vdcC gene, produced using pET22b(+) in E. coli BL21(DE3) at 25°C and 30°C with 0.1 mM IPTG for induction. Lane 1: soluble cell lysate, 25°C induction; Lane 2: insoluble protein, 25°C induction. Lane 3: soluble cell lysate, 30°C induction; Lane 4: insoluble protein, 25°C induction. The position of the vdcC expression product is denoted by an arrow. Molecular weight markers are shown on the left, in kilodaltons. 81

In hindsight, the strategy to express the vdcB and vdcC genes independently most likely would not have been successful, even if the proteins had been synthesized as soluble, active enzymes. As is detailed in Section 3.6.2 and Section 3.6.3 (below), it appears as though expression of at least the vdcC and vdcD genes, simultaneously, is required to produce vanillate decarboxylase activity.

3.6.2 Expression with pIJ702 in S. lividans 1326

Gene expression in Streptomyces is commonly performed using the high copy number plasmid, pIJ702 (Gusek & Kinsella, 1992). The vector, in combination with S. lividans

1326, has been used extensively to study antibiotic biosynthesis gene structure and expression. S. lividans 1326 carrying pKCSl acquired the ability to efficiently decarboxylate vanillic acid to guaiacol, while 5*. lividans 1326 carrying pKCS2 produced extremely low amounts of guaiacol (Figure 28). 5*. lividans 1326 wild type did not perform any detectable vanillic acid decarboxylation. These results suggest that transcription of the genes required for vanillic acid decarboxylation by S. lividans 1326 carrying pKCSl occurs from the constitutive mei promoter. There is possibly a low level of transcription from a natural promoter, however, as observed for S. lividans 1326 carrying pKCS2 that contains the gene cluster in the opposite orientation from the mei promoter (Figure 28 (b)).

3.6.3 Expression with pIJ680 in S. lividans 1326

The results of the aph promoter - vdc gene fusions are shown in Figure 30. S. lividans

1326 carrying pKCS3 (ydcBCD) or pKCS8 (vdcCD) converted vanillic acid to guaiacol 82 at approximately the same rate as the wild type strain, Streptomyces sp. D7. Sonicated cell-free extracts of S. lividans 1326 expressing the VDC system via pKCS3 exhibited decarboxylation of vanillic acid under both aerobic and anaerobic conditions (Figure 29).

These results confirm the involvement of the cloned gene products in a non-oxidative system.

Transcription of vdcC and vdcD, as demonstrated by pKCS8, results in vanillate decarboxylation by S. lividans 1326. Therefore, it is.the.vdcC gene that possibly encodes the active enzyme, and the 239 bp vdcD gene may encode a protein that is essential for enzyme stability or activation. The vdcD gene product of approximately 9 kDa seems too small to be the decarboxylase itself, as subunits for known aromatic acid decarboxylases range from 28 kDa to 66 kDa (Table 3).

3.7 Substrate specificity

Sonicated cell extracts of S. lividans 1326/pKCS8 were used to test the specificity of the

Streptomyces sp. D7 vanillate decarboxylase towards aromatic acids similar to vanillic acid. The following compounds were tested: 4-methoxy-3-hydroxybenzoate (isovanillic acid), 3,4-dimethoxybenzoate (veratrate), 3,4-dihydroxybenzoate (protocatechuate), 4- hydroxy-3,5-dimethoxybenzoate (syringate), 3,4,5-trihydroxybenzoate (gallate), 3- phenylpropenoate (frvms-cinnamate). The structures of these compounds are shown in

Figure 31. The UV spectrophotometry scans of assay mixtures at the time of substrate

addition, fifteen minutes post-addition and several hours post-addition were identical,

suggesting that the enzyme did not decarboxylate these compounds. By contrast, 83 ultraviolet spectrophotometry demonstrated vanillic acid decarboxylation to guaiacol after fifteen minutes in the soluble fractions of sonicated cell extracts, as shown in Figure

29. 84 A. 3.5

Time (h)

Figure 28: Decarboxylation of vanillic acid to guaiacol by recombinant Streptomyces lividans 1326 strains, (a) S. lividans 1326 carrying pKCSl; (b) S. lividans 1326 carrying pKCS2; (c) S. lividans 1326 wild type. Vanillic acid concentration is represented by squares (•); guaiacol concentration is represented by circles (•). 85

Figure 29: Results of incubating the insoluble and soluble fractions of sonicated cell extracts of 5". lividans 1326 - pKCS8 with 1 mM vanillic acid for fifteen minutes at 25°C under both aerobic (O2) and anaerobic (N2) conditions. These scans indicate that the decarboxylase activity is in the soluble fraction, and that the enzyme is active under both aerobic and anaerobic conditions. Similar experiments using other aromatic acid substrates did not demonstrate any biotransformation to decarboxylated derivatives. The

Xmax for guaiacol (275 nm) is indicated. 86

plJ680-aph VDC Activity

Paph

pKCS4 vdcB

Paph

pKCS5 ^ vdcC ;.:

Paph

pKCS6 ^ vdcD

Paph

pKCS7 ^ vdce _^cC

^ap/7 pKCS8 ^— ^cC vcfcD

' aph

pKCS3 \-— vdcB vdcC vdcD -

Figure 30: Results of placing the vdc genes under control of the aminoglycoside

phosphotransferase promoter (Paph) in pIJ680. 87

Substrate tested Structure Decarboxylation'

C02H

Vanillic acid YES

^^OCH3 OH

C02H

Isovanillic acid NO OH

CH3

C02H

Veratric acid NO

OCH3

OCH3

C02H

Syringic acid NO v H3CO'^j^ OCH3

C02H

S7 Gallic acid NO OH OH

C02H

f-cinnamic acid NO

Figure 31: Chemical structures of vanillic acid and similar aromatic acids used to test the substrate specificity of vanillate decarboxylase. 88

4. CONCLUSIONS

The goal of this study was to determine the genetic basis of the process of non-oxidative decarboxylation of vanillic acid in Streptomyces sp. D7. It was demonstrated that

Streptomyces sp. D7 performs the reaction by expressing products of a co-transcribed, three-gene cluster comprised of vdcB, vdcC and vdcD. The genes vdcC and vdcD encode proteins that catalyze the decarboxylation; vdcB is possibly involved in upstream reactions that convert more complex substrates to vanillic acid. Furthermore, the catabolic gene cluster appears to be controlled by the vdcA gene, which displays sequence similarity to the translated product of a putative positive transcriptional regulatory gene identified in the Streptomyces coelicolor A3 (2) genome sequencing project. This is, to my knowledge, the first report of the genes associated with the process of non-oxidative decarboxylation of aromatic acids in a microorganism. Further studies of this system should allow these genes to be incorporated into metabolically engineered microorganisms for the production of industrially useful chemical products such as guaiacol, catechol and adipic acid. 89

5. DISCUSSION

5.1 Aromatic acid non-oxidative decarboxylases are multi-subunit

enzymes

Although there are many reports of microbial non-oxidative decarboxylases active on aromatic acids in the literature, only recently have enzyme purifications been successful, as these proteins appear to be unstable in cell-free extracts. Of the enzymes purified thus far, all share one common feature: they are single polypeptides, which.form multi-subunit enzyme complexes. However, depending on substrate and organism, the molecular mass of the polypeptide, as well as the number of subunits, is variable. Several examples of aromatic acid non-oxidative decarboxylases and their subunit configurations are listed in

Table 3. I have demonstrated that Streptomyces sp. D7 produces a protein of approximately 52 kDa when grown in the presence of vanillic acid, and suggest that two additional functionally related proteins of 36 kDa and 9 kDa are also synthesized. The exact functions of these proteins remain unknown, although it appears that vdcC encodes the vanillic acid decarboxylase. Experiments in which vdcC and vdcD were expressed under the control of the aph promoter in pIJ680 resulted in vanillate decarboxylase activity, and it is possible that the 52 kDa VdcC is a subunit of a complex similar to non- oxidative decarboxylases described in the literature (Table 3). The amino-terminus of the

VdcC protein is highly similar to the limited amino acid sequence obtained from the purified /?-hydroxybenzoate carboxy-lyase of Clostridium hydroxybenzoicum. The

Clostridium enzyme was purified and characterized, but only limited amino acid

sequence was obtained. The enzyme is responsible for decarboxylation of p-

hydroxybenzoate to phenol, a dead-end metabolite. With the exception of the amino acid

sequence similarity between the vanillic acid induced protein of Streptomyces sp. D7 and 90

Enzyme Organism Configuration Reference 4-hydroxybenzoate Clostridium 350 kDa (6 subunits of 57 kDa) He and decarboxylase hydroxybenzoicum Wiegel, 1995

3,4-dihydroxybenzoate Clostridium 270 kDa (5 subunits of 57 kDa) He and decarboxylase hydroxybenzoicum Wiegel, 1996

4,5 -dihydroxyphthalate Pseudomonas testosteroni 150 kDa (4 subunits of 38 kDa) Nakazawa decarboxylase and Hayashi, 1978

4,5 -dihydroxyphthalate Pseudomonas fluorescens 420 kDa (6 subunits of 66 kDa) Pujar and decarboxylase Gibson, 1985

2,3 -dihydroxybenzoate Aspergillus niger 120 kDa (4 subunits of 28 kDa) Kamath et decarboxylase al, 1987

2,3 -dihydroxybenzoate Trichosporon cutaneum 66.1 kDa (2 subunits of 36.5 Anderson and decarboxylase kDa) Dagley, 1981

3,4,5- Pantoea agglomerans T71 320 kDa (6 subunits of 57 kDa) Zeida et al, trihydroxybenzoate 1998 decarboxylase

Table 3: Variations in subunit size and configuration - characteristics of some microbial aromatic acid non-oxidative decarboxylases. 91 the C. hydroxybenzoicum enzyme, it is difficult to assign the exact function(s) of the vdcC or vdcD gene products in the reaction. Neither polypeptide shows a relationship to any characterized enzymes in the databases, although the amino terminal sequence of the

C. hydroxybenzoicum enzyme was noted to have weak similarity to uroporphyrinogen decarboxylase of Synechococcus sp (He & Wiegel, 1995). However, extending this comparison to the Streptomyces sp. D7 protein would be speculative, as the

Synechococcus sp. uroporphyrinogen decarboxylase amino-terminal sequence bears almost no resemblance to the Streptomyces sp. D7 protein. The product of vdcB has primary amino acid sequence highly similar to phenylacrylate decarboxylase (PAD) from yeast. In light of this functional implication, it may be possible that the vdcB product is involved in transformation of acrylic phenolic compounds to substrates suitable for downstream gene products, possibly VdcC. The putative trans-membrane region noted in the yeast PAD GenBank database entry is highly conserved among the bacterial PAD homologues' amino-terminal regions (Figure 32). Therefore, it is possible that the product of the vdcB gene is membrane associated. Further biochemical research is needed to elucidate the true roles of the components of the VDC gene cluster.

5.2 Distribution of the VDC gene cluster among streptomycetes

The results of the Southern hybridization experiments demonstrate that the homologues of the VDC cluster genes are also apparently present in the genome of S. setonii 75Vi2.

This observation reinforces the notion that the gene cluster described in this thesis is responsible for vanillate decarboxylation (recall that S. setonii 75Vi2 is the streptomycete well characterized in the literature for vanillate decarboxylation). Of thirteen 92 streptomycetes screened in similar Southern blots (results not shown), only three - S. setonii 75Vi2, Streptomyces sp. D7, and Streptomyces sp. 2065 (another environmental isolate from our collection) appeared to carry these genes. This observation suggests that the ability to decarboxylate vanillic acid to guaiacol is not a widespread trait among streptomycetes. Degradation via demethylation to protocatechuate, by the enzyme vanillate demethylase, is likely the more common route for catabolism of vanillic acid in streptomycetes, as well as other prokaryotes. In fact, a cosmid olthe S. coelicolor A3(2) genome sequencing project contains homologues of the vanillate demethylase genes from

Pseudomonas sp. It was previously demonstrated, by the colorimetric Rothera assay, that protocatechuate 3,4-dioxygenase is induced in Streptomyces sp. D7 in the presence of vanillic acid, suggesting that the initial attack on vanillic acid in this organism is performed by vanillate demethylase as well as vanillate decarboxylase (Chow, 1996).

However, it is apparent from the current study that vanillate decarboxylase is very highly expressed in Streptomyces sp. D7, as evidenced by the almost equimolar conversion of vanillic acid to guaiacol by Streptomyces sp. D7 mycelia. If vanillate demethylase is synthesized in response to vanillate, the levels of that enzyme are likely very low relative to vanillate decarboxylase.

5.3 Substrate specificity

The apparent high substrate specificity of the VDC system is perhaps not surprising in light of other non-oxidative decarboxylase studies. Most decarboxylases studied thus far have been shown to be very specific for one substrate, although several are active against two related aromatic acids. For example, gallate decarboxylase from Pantoea 93 agglomerans T71 is highly substrate specific (Zeida et al, 1998), while p- hydroxybenzoate carboxy-lyase from Clostridium hydroxybenzoicum is active against both/>-hydroxybenzoate and protocatechuate (He & Wiegel, 1995). Klebsiella aerogenes was biochemically demonstrated to produce a number of non-oxidative decarboxylases, each enzyme specific for a different aromatic acid substrate (Grant & Patel, 1969). It will be interesting in future studies to determine the catalytic sites of these enzymes, in order to elucidate which amino acid residues affect substrate binding.

5.4 Primary structure motifs

Microbial non-oxidative decarboxylase systems reported in the literature (Grant & Patel,

1969; Yoshida & Yamada, 1985; Nakajima et al, 1992; Huang et al, 1993; Santha et al,

1995; He & Wiegel, 1995; He & Wiegel, 1996; Zeida et al., 1998) all have minimal or no requirements of cofactors for activity. The vanillate decarboxylase system of

Streptomyces sp. D7 appears active in the absence of oxygen. Consistent with this, amino acid sequence analysis of all three VDC gene products failed to reveal the presence of any cofactor binding motifs, including NAD and FAD, which are characteristic of oxidative enzymes.

5.5 Transcriptional activation

The proteomics analyses presented previously (Chow, 1996), combined with the genetic analysis provided in this study, together demonstrate that genes encoding proteins linked to vanillic acid decarboxylation are induced, directly or indirectly, by vanillic acid itself.

This observation is supported by the northern blot data, indicating that the VDC gene 94 cluster is transcribed in one polycistronic mRNA product upon induction by vanillic acid.

These results suggest that transcription of the gene cluster is under tight regulatory control. In fact, preliminary nucleotide sequence analysis upstream of the VDC gene cluster revealed a divergent putative regulatory gene. Future work will complete the characterization of the putative regulatory gene and allow elucidation of its relationship to the VDC gene cluster. The mechanism by which vanillic acid enters streptomycete mycelia and activates transcription of the VDC gene cluster should prove to be an interesting model of environmental sensing and response. Streptomyces has a high proportion of two component signal transduction regulators. Recent data from the S. coelicolor A3(2) genome project predicts a total of 160 two-component regulators, approximately double the number found in any other genome thus far analyzed (D.

Hopwood, personal communication).

5.6 Biodegradation — a result of the microbial community gene pool

The wide distribution of homologues of the VDC genes throughout the microbial world, mostly encoded by chromosomes, but also plasmid-borne as in the case of pNLl in

Sphingomonas aromaticivorans, suggests that these gene products provide useful metabolic abilities for their hosts. However, non-oxidative decarboxylation of most aromatic acids yields toxic phenolic compounds and, in many cases cited in the literature, the microorganisms do not possess appropriate mechanisms to further degrade the compounds produced by these dead-end pathways.. For example, in this study,

Streptomyces sp. D7 was able to rapidly convert vanillic acid to guaiacol, but was unable to further degrade the guaiacol. In another example, Clostridium hydroxybenzoicum decarboxylated j?-hydroxybenzoate to phenol, and protocatechuate to catechol without 95 further metabolism. Non-oxidative decarboxylation remains an enigma of microbial metabolism (Frost & Draths, 1995) and in the case of the C. hydroxybenzoicum enzymes,

He and Weigel mention that "a direct metabolic function in the bacterium is not known for either of the two enzymes" (He & Weigel, 1996). The functions of these seemingly toxic metabolic reactions of microorganisms are not readily apparent; however, in natural ecosystems, it can be imagined that other organisms in a consortium would mineralize and remove these toxins from the environment. For example, Streptomyces setonii 75Vi2

(Pometto III et al, 1981; Sutherland, 1986) and a Moraxella sp. (Sterjiades, et al, 1982) were demonstrated to demethylate methoxylated aromatic compounds such as guaiacol to catechol, leading to subsequent mineralization. Evidence implicating cytochrome P-450 systems was provided in both cases. The preliminary sequence data we have obtained, indicating that a cytochrome P-450 gene is located within a few kilobases of the VDC gene cluster in 5*. setonii 75Vi2, suggests that vanillate decarboxylation to guaiacol is associated with demethylation of guaiacol to catechol in that organism. S. setonii 75Vi2 was also observed to degrade phenol (Antai and Crawford, 1982), adding to its reputation as a catabolically diverse organism. Streptomyces sp. D7 does not appear to be as catabolically diverse as S. setonii 75Vi2, and from additional sequencing studies of the ends of BamHl subclones of the 13 kb genomic library clone, does not appear to possess

a cytochrome P-450 gene nearby the VDC gene cluster. This observation is in

accordance with the fact that the strain does not metabolize guaiacol - that is, guaiacol is

a dead end metabolite. One could speculate that while S. setonii 75Vi2 possesses a

complete complement of lignin-related aromatic acid catabolic genes, Streptomyces sp.

D7 acquired (from another streptomycete or microorganism) only a partial set. Indeed, 96 there is a growing focus on biodegradation, not from the standpoint of an individual microorganism, but rather as a coordinated function of the entire gene pool (Wackett,

1999). Efficient biodegradation is therefore likely the result of natural consortiums of microorganisms. The observation that Streptomyces sp. D7 converts vanillic acid to the toxic guaiacol, but is unable to remove the guaiacol from its environment appears to

support this view. However, a microorganism producing metabolites toxic to itself defies the principles of natural selection. The observation of the production of vast amounts of

guaiacol from vanillic acid by Streptomyces sp. D7 is based on laboratory experiments, in which unnaturally high concentrations of vanillic acid were fed to the microorganism. In

contrast, natural settings such as forest soils most likely do not have high amounts of vanillic acid freely available to the microorganisms. The release of such low molecular weight aromatic compounds from lignin is a rate-limited process, determined by the

activity of fungi and abiotic processes. In addition, guaiacol produced by organisms such

as Streptomyces sp. D7 would most likely be further degraded by other organisms in the environment. Therefore, Streptomyces sp. D7 would not be living with toxic levels of

guaiacol, even if its vanillate decarboxylase system were a result of the acquisition or

deletion of certain catabolic genes from its genome, leading to incomplete degradation of vanillic acid.

5.7 Sphingomonas aromaticivorans F199 catabolic plasmid pNLl: filling

in the missing links

Recent publication of the entire sequence of the 184 kb catabolic plasmid pNLl of

Sphingomonas aromaticivorans F199 (Romine et al, 1999) sheds additional light on the

role of the VDC gene cluster in biodegradation processes. pNLl is a conjugative 97 plasmid, which contains 186 open reading frames, 70 of which are likely associated with catabolism or transport of aromatic compounds. On this plasmid are 15 gene clusters encoding biodegradative enzymes. Among genes encoding biphenyl and p-cresol degradative enzymes, lie homologues to vdcB, vdcC and vdcD. Amino acid sequence alignments between the Streptomyces sp. D7 vdcB and vdcC genes and the

Sphingomonas homologues are shown in Figure 27. Between the vdcB and vdcC homologues, there are two ORFs not found in Streptomyces sp. D7: pchFa (p-cresol methylhydroxylase) and vdh (vanillin oxidoreductase).

Figure 32 (following pages): Amino acid sequence comparisons between the Streptomyces sp. D7 vdcB translation product (strPAD) and its Sphingomonas pNLl homologue (sphPAD) (a), and the Streptomyces sp. D7 vdcC translation product (strC) and its Sphingomonas pNLl homologue (sphC) (b). 98

(a) 1 50 sphPAD MKRMWGITG ATGSVYGLRL LELLRETGGW ETHLVMSPAA LLNIREELPE strPAD M.RLVVGMTG ATGAPFGVRL LENLRQLPGV ETHLVLSRWA RTTIELETGL Consensus M-R-VVG-TG ATG G-RL LE-LR G- ETHLV-S--A 1—E

51 100 sphPAD GKARLEALAD VVHNVRNVGA SIASGSFVCE GMAIAPCSMR TLGAVAHALS strPAD SVAEVSALAD VTHHPEDQGA TISSGSFRTD GMVIVPCSMK TLAGIRTGYA Consensus —A ALAD V-H GA -I-SGSF GM-I-PCSM- TL

101 150 sphPAD DNLITRAADV MLKERRRLVM ITREAPLNLA HLRNMTACTE MGAVIFPPVP strPAD EGLVARAADV VLKERRRLVL VPRETPLSEI HLQNMLELAR MGVQLVPPMP Consensus —L—RAADV -LKERRRLV- —RE-PL HL-NM MG PP-P

151 200 sphPAD AFYARPTSLA DVVDHTCMRV LDLFGLHAKS EKRWQGLSKE AASLVPGAGQ strPAD AFYNNPQTVD DIVDHVVARI LDQFDLPAPA ARRWAGMRAA RAAARSFGDA Consensus AFY — P D-VDH R- LD-F-L-A— —RW-G -A

201 sphPAD MEGN strPAD A Consensus 99

b

1 50 StrC MAYDDLRSFL DTLEKEGQLL RITDEVLPEP DLAAAANATG sphC MTMNDLPNRA RSISSLRDFL ELLEDAGQAI TWSDAVMPEP GVRNIAVAAS Consensus LR-FL —LE—GQ— D-V-PEP A-A—

51 100 strC RIGENAPALH FDNVKGFTDA RIAMNVHGSW ANHALALGLP KNTPVKEQVE sphC RDANGAPAIV FDNITGYPGK RLAVGVHGSW DNIALLLGRP KGTTIRELFF Consensus R APA— FDN—G R-A—VHGSW -N-AL-LG-P K-T E

101 150 strC EFARRW..DA FPVAPERREE APWRENTQEG EDVDLFSVLP LFRLNDGDGG sphC EIAGRWGDQE AQISFVPEAQ APVHE.CRIE QDINLYDVLP VYRINEYDGG Consensus E-A-RW AP—E -D—L—VLP —R-N—DGG

151 200 strC FYLDKAAVVS RDPEDRDDFG KQNVGTYRIQ VIGTNRLAFH PA.MHDVAQH sphC FYIGKASVAS RDPLDPDNFG KQNVGIYRLQ IQGPDTFTLM TIPSHDMGRQ Consensus FY—KA-V-S RDP-D-D-FG KQNVG-YR-Q —G HD

201 250 strC LRKAEEKGED LPIAITLGND PVMAIVAGMP MAYDQSEYEM AGALRGAPAP sphC IMAAEREGVP LKIAVMLGNH PGLAAFAATP IGYEESEYSY ASAMMGAPIR Consensus AE — G— L-IA—LGN- P—A—A—P —Y—SEY— A-A—GAP—

251 300 strC IATAPLTGFD VPWGSEVVIE GVIESRKRRI EGPFGEFTGH YSGGRRMPVI sphC LTKSG.NGID ILADSEIVIE AELQPGGREL EGPFGEFPGS YSGVRKAPIF Consensus G-D SE-VIE EGPFGEF-G- YSG-R—P—

301 350 strC RVERVSYRHE PVFESLYLGM PWNECDYLVG PNTCVPLLKQ LRAEFPEVQA sphC KVTAVSHRRD PIFENIYIGR GWTEHDTLIG LHTSAPIYAQ LRQSFPEVTA Consensus -V—VS-R— P-FE—Y-G- -W-E-D-L-G — T — P Q LR—FPEV-A

351 400 strC VNAMYTHGLM VIISTAKRYG GFAKAVGMRA MTTPHGLGYV AQVILVDEDV sphC VNALYQHGLT GIISVKNRMA GFAKTVALRA LSTPHGVMYL KNLIMVDADV Consensus VNA-Y-HGL- -IIS R— GFAK-V—RA —TPHG—Y- I-VD-DV

401 450 strC DPFNLPQVMW AMSAKVNPKD DVVVIPNLSV LELAPAAQPA GISSKMIIDA sphC DPFDLNQVMW ALSTRTR.AD DIIVLPNMPA VPIDPSAVVP GKGHRLIIDA Consensus DPF-L-QVMW A-S D D—V-PN P-A G IIDA

451 489 strC TTPVAPDVRG NFSTPAKDLP ETAEWAARLQ RLIAARV— sphC TSYLPPDPVG EAHLVTPPTG DEIDALSKRI REMQLGALS Consensus T PD—G 100

The pchFa gene was originally isolated from Pseudomonas putida NCIMB 9866, and encodes an enzyme that oxidizes the methyl group of /?-cresol to an aldehyde. The vdh gene was originally isolated from Pseudomonas fluorescens AN 103 (Walton, Genbank,

1997), and encodes an NAD+-dependent oxidoreductase, which converts the aldehyde vanillin to vanillic acid. These genes are followed by homologues to vdcC and vdcD, for which this thesis gives evidence (pKCS7 in S. lividans 1326) that the gene products in

Streptomyces sp. D7 are involved in vanillic acid non-oxidative decarboxylation. The location of the vdc gene homologues in relation to other genes on pNLl is shown in

Figure 33. From this one can create a catabolic scenario for these genes on plasmid pNLl, that phenylacrylate derivatives are transformed to aromatic aldehydes, which are oxidized to aromatic acids, then non-oxidatively decarboxylated to catechol derivatives for mineralization via catechol dioxygenase ring cleavage enzymes. In fact, the meta- cleavage enzyme catechol 2,3-dioxygenase is found on pNLl. It should be noted that the translation products of the pchFa, vdh, vdcB and vdcC homologues on pNLl were approximately 55% identical to their Streptomyces sp. D7 and Pseudomonas spp. counterparts, and therefore may be similar in function but not substrate specificity. For example, the pNLl genes may encode non-oxidative, decarboxylase enzymes that transform toluene derivatives instead of lignin-related aromatic compounds.

Streptomyces sp. D7 did not contain the aldehyde oxidoreductase gene found on pNLl, and therefore would not be expected to carry through the bioconversion of phenylacrylates to aromatic acids. Indeed, & lividans 1326 carrying pKCS8 effectively

decarboxylated vanillic acid to guaiacol, no observable biotransformation activity was noted when the organism was exposed to ferulic acid, the acrylate derivative of vanillic 101 acid. The role of the VDC gene cluster for Streptomyces sp. D7 in nature remains to be determined, and one issue that arises from this study is whether or not the organism acquired the cluster through horizontal gene transfer events during evolution. Standard plasmid isolation procedures performed on Streptomyces sp. D7 did not reveal the presence of any plasmids, but this issue should be investigated further in future studies, perhaps using pulsed field gel electrophoresis techniques. Another question is whether the organism originally carried all the genes necessary for vanillic acid mineralization, but lost some downstream catabolic genes (such as those for the degradation of guaiacol) through deletion events. Finally, it is possible that the VDC gene cluster encodes enzymes for which nature did not intend to decarboxylate vanillic acid, but rather another substrate not tested in the work presented here. Streptomyces sp. D7 could be able to degrade a yet to be determined complex substrate with a vanillic acid-like degradation intermediate, which is processed by the genes described in this thesis work; 102

B.

% identity to pNL1 ORF Functional description of closest relative 8. subtilis ORF

orf1244 Phenylacrylic acid decarboxylase homolog 50 orf1272 conserved hypothetical protein 47 orf1280 conserved hypothetical protein 32

FIGURE 33: pNLl of Sphingomonas aromaticivorans. (a) Physical map of pNLl and the location of the VDC gene cluster homologues; (b) listing of the pNLl open reading frames similar to ORFs from the Bacillus subtilis genome (From Romine, et al., 1999) 103

5.8 Metabolic Engineering Applications for Vanillate Decarboxylase

The elaborate metabolic engineering scheme described for E. coli AB2834, which enables the production of adipic acid using D-glucose as a starting material, is an example of what can be achieved by mixing and matching catabolic genes in an appropriate host microorganism (Draths & Frost, 1994).

Similarly, the genes encoding the components of the vanillate decarboxylase system of

Streptomyces sp. D7 could be used in engineered systems for the transformation of lignin-rich plant and timber waste to useful commodity or fine chemicals. Engineering such a system would probably not be as complex as the development of E. coli AB2834 for conversion of D-glucose to adipic acid. The by-products of wood pulp and other plant manufacturing processes, such as olive oil production (Ramos, J., personal communication), consist of large amounts of methoxylated aromatic chemicals. In theory, these aromatic chemicals could be separated from the waste stream and refined to industrially useful chemicals such as catechol and cis,cis-muconic acid (for drug syntheses), adipic acid (for nylon), or guaiacols (for medicines and scent compounds).

To enable this process, a microorganism would have to be modified with a suite of genes encoding enzymes to produce the desired end products. For the refining of vanillate, a microorganism expressing vanillate decarboxylase would be an efficient producer of guaiacol, while a strain expressing genes for vanillate decarboxylase and guaiacol- demethylating Cytochrome P-450 (cloned and sequenced from Streptomyces setonii

75Vi2 by M.K. Pope and S. Baily in this laboratory) would produce catechol. If this catechol producing strain is supplemented with the gene for catechol 1,2-dioxygenase, 104 adipic acid could be produced. A diagram of this proposed scheme is shown in Figure

34. With the current low prices for petroleum products, commercialization of such biorefining technology is not practical as the desired products can be inexpensively produced from benzene. However, with the continually changing supply and demand of petroleum, and with increasingly stringent environmental regulations being imposed on industry, biorefining technology may become not only practical, but necessary. 105

COOH

Lignin solubilization product

VANiLLATE DECARBOXYLASE

Used for fragrances and medicines

Used as a drug synthesis precursor

Used as a drug synthesis precursor HOOC-

FIGURE 34: Vanillate decarboxylase, Cytochrome P-450 and catechol 1,2-dioxygenase can be used in combinations to produce various industrially useful chemicals from vanillic acid as a starting material. 106

5.9 Future directions

The cloning of the genes required for vanillic acid non-oxidative decarboxylation, as presented in this work, will allow us to expand our knowledge of the reaction.

Purification of active recombinant protein products encoded by the gene cluster will enable detailed reaction kinetics to be determined for the decarboxylation process.

Future studies will shed light on the structure and catalytic function of vanillic acid decarboxylase from Streptomyces sp. D7. It will be interesting to determine if the numerous sequence homologues from microbial genome databases are indeed other non- oxidative decarboxylases. The transcriptional regulation of the VDC gene cluster in response to environmental stimuli will be another interesting aspect to investigate.

Accordingly, one experiment planned for the near future will be to clone the entire VDC gene cluster, including the divergent putative regulatory gene, into a suitable plasmid expression vector for expression in S. lividans 1326. It has already been demonstrated by

2D-PAGE and mRNA analyses that in Streptomyces sp. D7, the vanillate decarboxylase gene cluster is inducible in response to vanillic acid. A recombinant S. lividans 1326 strain carrying the VDC gene cluster and the associated regulatory gene should also demonstrate vanillic acid inducibility, unless the regulatory system requires other components not present in S. lividans 1326. Non-oxidative decarboxylases represent connections between the two major branches of aromatic acid catabolism, characterized by either catechol or protocatechuate central intermediates (Figure 35). By joining these two pathways, this class of enzymes supports the concept of aromatic catabolism as being a web of interconnecting biodegradative processes, and not separate, distinct pathways

(Crawford and Olson, 1978). From an applied standpoint, knowledge gained from such 107

Lignin

C02H VDC

T OCH3 OH OH vanillate guaiacol

VanA, VanB Cytochrome P-

C02H PDC

, OH OH OH protocatechuate catechol PcaG, PcaH CatA

mineralization mineralization

FIGURE 35: Non-oxidative decarboxylases represent connections between the two major branches of aromatic acid catabolism, characterized by either catechol or protocatechuate central intermediates. VDC = vanillate decarboxylase; PDC = protocatechuate decarboxylase; VanA, VanB = vanillate demethylase; PcaG, PcaH = protocatechuate 3,4-dioxygenase; CatA = catechol 1,2-dioxygenase 108 endeavors should ultimately allow this class of enzyme to be used for a variety of metabolic engineering applications. Currently, the world's major chemical companies are refocusing their business strategies towards producing many bulk commodity chemicals via microbial fermentations (Alper, 1999). Genetic engineering of E. coli has already proven that aromatic acid decarboxylases are valuable tools for biomass conversion pathway development. Vanillate decarboxylase could be useful as a component in the engineered bioconversion of complex lignin • molecules, through intermediates such as ferulic acid, down to industrially useful low molecular weight compounds such as catechol, adipic acid (Frost & Draths, 1995) and cis.cis-muconic acid

(Yoshikawa et al, 1990). 109

6. LITERATURE CITED

Alper, J. (1999). Engineering metabolism for commercial gains. Science 283, 1625-

1626.

Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990). Basic local alignment search tool. J Mol Biol 215, 403-410.

Anderson, J.J. & Dagley, S. (1981). Catabolism of tryptophan, • anthranilate, and 2,3- dihydroxybenzoate in Trichosporon cutaneum. J Bacteriol 146, 291-297.

Antai, S.P. & Crawford, D.L. (1983). Degradation of phenol by Streptomyces setonii.

Can J Microbiol 29, 142-143.

Atlas, R. & Bartha, R. (1993). Survey of microorganisms, p. 506-532. In Microbial

Ecology: Fundamentals and Applications. Benjamin/Cummings, Redwood City.

Bibb, M.J., Ward, J.M. & Hopwood, D.A. (1978). Transformation of plasmid DNA into

Streptomyces protoplasts at high frequency. Nature 274, 398-400.

Brunei, F. & Davison, J. (1988). Cloning and sequencing of Pseudomonas genes encoding vanillate demethylase. J. Bacteriol 170, 4924-4930. no

Chow, K (1996). Two-dimensional polyacrylamide gel electrophoretic analysis of protein synthesis during aromatic acid catabolism by Streptomyces violaceusniger.

M.Sc. thesis.

Chow, K.T., Pope, M.K. & Davies, J. (1999). Characterization of a vanillic acid non- oxidative decarboxylation gene cluster from Streptomyces sp. D7. Microbiology 145,

2393-2403.

Crawford, R.L. & Olson, P.R. (1978). Microbial catabolism of vanillate: decarboxylation to guaiacol. Appl and Environ Microbiol 36, 539-543.

Draths, K.M. and Frost, J.W. (1994). Environmentally compatible synthesis of adipic acid from D-glucose. J. Am. Chem. Society 116, 399-400.

Frost, J.W. & Draths, K.M. (1995). Biocatalytic syntheses of aromatics from D-glucose: renewable microbial sources of aromatic compounds. Ann. Rev. Microbiol. 49, 557-79.

Garrels, J.I. (1979). Two dimensional gel electrophoresis and computer analysis of proteins synthesized by clonal cell lines. J Biol Chem 254, 7961-7977.

Grant, D.J.W. & Patel, J.C. (1969). The non-oxidative decarboxylation of p- hydroxybenzoic acid, gentisic acid, protocatechuic acid and gallic acid by Klebsiella aerogenes (Aerobacter aerogenes). Antonie Leeuwenhoek 35, 325-343. Ill

Grund, E., Knorr, C. & Eichenlaub, R. (1990). Catabolism of benzoate and monohydroxylated benzoates by Amycolatopsis and Streptomyces spp. Appl and Environ

Microbiol 56, 1459-1464.

Gusek, T.W. & Kinsella, J.E. (1992). Review of the Streptomyces Uvidans/vQcXox pIJ702 system for gene cloning. Crit. Rev. Microbiol. 18, 247-260.

He, Z. & Wiegel, J. (1995). Purification and characterization of an oxygen-sensitive reversible 4-hydroxybenzoate decarboxylase from Clostridium hydroxybenzoicum. Eur J

Biochem 229, 77-82.

He, Z. & Wiegel, J. (1996). Purification and characterization of an oxygen-sensitive, reversible 3,4-dihydroxybenzoate decarboxylase from Clostridium hydroxybenzoicum. J

Bacteriol 178,3539-3543.

Hopwood, D.A. personal communication.

Hopwood, D.A., Bibb, M.J., Chater, K.F., Kieser, T., Bruton, C.J., Kieser, H.M., Lydiate,

D.J., Smith, CP., Ward, J.M. & Schrempf, H. (1985). Genetic manipulation of

Streptomyces: a laboratory manual. The John Innes Foundation, Norwich. 112

Huang, Z., Dostal, L. & Rosazza, J.P.N. (1993). Mechanisms of ferulic acid conversions to vanillic acid and guaiacol by Rhodotorula rubra. J Biol Chem 268, 23954-23958.

Kamath, A.V., Dasgupta, D. & Vaidyanathan, CS. (1987). Enzyme-catalyzed non- oxidative decarboxylation of aromatic acids: I. Purification and spectroscopic properties of 2,3-dihydroxybenzoic acid decarboxylase from Aspergillus niger. Biochem Biophys

Res Comm 145, 586-595.

Katz, E., Thompson, C.J. & Hopwood, D.A. (1983). Cloning and expression of the tyrosinase gene from Streptomyces antibioticus in Streptomyces lividans. J Gen

Microbiol 129, 2703-2714.

Kirby, K.S., Fox-Carter, E. & Guest, M. (1967). Isolation of deoxyribonucleic acid and ribosomal ribonucleic acid from bacteria. Biochem J 104, 258-262.

Kirk, T.K. (1987). Enzymatic "combustion": the microbial degradation of lignin. Ann

Rev Microbiol 41, 465-505.

Nakajima, H., Otani, C. & Niimura, T. (1992). Decarboxylation of gallate by cell-free extracts of Streptococcus faecalis and Klebsiella pneumoniae isolated from rat feces. J

Food Hyg Soc Japan 33, 371-376. 113

Nakazawa, T. & Hayashi, E. (1978). Phthalate and 4-hydroxyphthalate metabolism in

Pseudomonas testosteroni: purification and properties of 4,5-dihydroxyphthalate decarboxylase. Appl Environ Microbiol 36, 264-269.

Nishikawa, S., Sonoki, T., Kasahara, T., Obi, T., Kubota, S., Kawai, S., Morohoshi, N. &

Katayama, Y. (1998). Cloning and sequencing of the Sphingomonas (Pseudomonas) paucimobilis gene essential for the O demethylation of vanillate and syringate. Appl and

Environ Microbiol 64, 836-842.

Oh, S.H. and Chater, K.F. (1997). Denaturation of circular or linear DNA facilitates targeted integrative transformation of Streptomyces coelicolor A3 (2): possible relevance to other organisms. J Bacteriol 179(1):122-127.

Oritz, M.L., Calero, M., Patron, C.F., Castellanos, L & Mendez, E. (1992). Imidazole-

SDS-Zn reverse staining of proteins in gels containing or not SDS and microsequence of individual unmodified electroblotted proteins. FEBS Lett 296, 300-304.

Pometto III, A.L., Sutherland, J.B. & Crawford, D.L. (1981). Streptomyces setonii: catabolism of vanillic acid via guaiacol and catechol. Can J Microbiol 27, 636-638.

Pujar, B.G. & Gibson, D.W. (1985). Phthalate metabolism in Pseudomonas fluorescens

PHK: purification and properties of 4,5-dihydroxyphthalate decarboxylase. Appl

Environ Microbiol 49, 374-376. 114

Romine, M.F., Stillwell, L.C., Wong, K.-K., Thurston, S.J., Sisk,.E.C, Sensen, C.W.,

Gaasterland, T., Saffer, J.D., Fredrickson, J.K. & Saffer, J.D. (1999). Complete sequence of a 184 kb catabolic plasmid from Sphingomonas aromaticivorans strain F199. J.

Bacteriol 181, 1585-1602.

Santha, R., Savithri, H.S., Rao, A. & Vaidyanathan, CS. (1995). 2,3-dihydroxybenzoic acid decarboxylase from Aspergillus niger. A novel decarboxylase. Eur J Biochem 230,

104-110.

Segura A. & Ornston, N.L. (1997). pZR135: vanillate demethylase region in

Acinetobacter. NCBI Entrez database submission (unpublished), accession AF009672.

Sterjiades, R., Sauret-Ignazi, G., Dardas, A. & Pelmont, J. (1982). Properties of a bacterial strain able to grow on guaiacol. FEMS Microbiol Lett 14, 57-60.

Sutherland, J.B., Crawford, D., & Pometto III, A.L. (1981). Catabolism of substituted benzoic acids by Streptomyces species. Appl and Environ Microbiol 41, 442-448.

Sutherland, J.B. (1986). Demethylation of veratrole by cytochrome P-450 in

Streptomyces setonii. Appl and Environ Microbiol 52, 98-100. 115

Thomas, L & Crawford, D.L. (1998). Cloning of clustered Streptomyces viridosporus

T7A lignocellulose catabolism genes encoding peroxidase and endoglucanase and their extracellular expression in Pichia pastoris. Can J. Microbiol 44, 364-372.

Thompson, C.J., Ward, J.M. & Hopwood, D.A. (1980). DNA cloning in Streptomyces: resistance genes from antibiotic-producing species. Nature 286, 525-527.

Thompson, C.J., Ward, J.M. & Hopwood, D.A. (1982). Cloning of antibiotic resistance and nutritional genes in streptomycetes. J Bacteriol 151, 668-677.

Wackett, L.P., Ellis, L.B.M., Speedie, S.M., Hershberger, CD., Knackmuss, H-J.,

Spormann, A.M., Walsh, C.T., Forney, L.J., Punch, W.F., Kazic, T., Kanehisa, M. &

Berndt, D.J. (1999). Predicting microbial biodegradation pathways. ASM News 65, 87-

93.

Walton, N.J. (1998). Pseudomonas fluorescens genes encoding p-hydroxycinnamoyl

CoA hydratase/lyase and vanillin: NAD+ oxidoreductase (unpublished). GenBank accession Y13067.

Yoshida, H. & Yamada, H. (1982). Microbial production of pyrogallol through decarboxylation of gallate. Agric Biol Chem 49, 659-663. 116

Yoshikawa, N., Mizuno, S., Ohta, K. & Suzuki, M. (1990). Microbial production of cz'^cw-muconic acid. J Biotechnol 14, 203-210.

Zeida, M., Wieser, M., Yoshida, T., Sugio, T. & Nagasawa, T. (1998). Purification and characterization of gallic acid decarboxylase from Pantoea agglomerans T71. Appl and

Environ Microbiol 64,4743-4747.